Package 'stevemisc'

Description

%nin% finds non-matching elements in a given vector. It is the negation of %in%.

Usage

a %nin% b

Arguments

Details

This is a simple negation of %in%. I use it mostly for columns in a data frame.

Value

%nin% finds non-matching elements and returns one of two things, depending on the use. For two simple vectors, it will report what matches and what does not. For comparing a vector within a data frame, it has the effect of reporting the rows in the data frame that do not match the supplied (second) vector.

Examples

library(tibble)
library(dplyr)

# Watch this subset stuff

dat <- tibble(x = seq(1:10), d = rnorm(10))
filter(dat, x %nin% c(3, 6, 9))

Description

binred_plot() provides a diagnostic of the fit of the generalized linear model by "binning" the fitted and residual values from the model and showing where they may fall outside 95% error bounds.

Usage

binred_plot(model, nbins, plot = TRUE)

Arguments

Details

The number of bins the user wants is arbitrary. Gelman and Hill (2007) say that, for larger data sets (n >= 100), the number of bins should be the rounded-down square root of the number of observations from the model. For models with a number of observations between 10 and 100, the number of bins should be 10. For models with fewer than 10 observations, the number of bins should be the rounded-down number of observations (divided by 2). The default is the rounded square root of the number of observations in the model. Be smart about what you want here.

Value

bindred_plot() returns a plot as a ggplot2 object, as a default. The y-axis is the mean residuals of the particular bin. The x-axis is the mean fitted values from the bin. Error bounds are 95%. A LOESS smoother is overlaid as a solid blue line.

If plot = FALSE, the function returns a data frame of the binned residuals and a summary about whether the residuals are in the error bounds.

Author(s)

Steven V. Miller

Examples

M1 <- glm(vs ~ mpg + cyl + drat, data=mtcars, family=binomial(link="logit"))

binred_plot(M1)

Description

This recodes a numeric vector, character vector, or factor according to fairly simple recode specifications that former Stata users will appreciate. Yes, this is taken from John Fox's recode() unction in his car package. I'm going with carrec() (i.e. shorthand for car::recode(), phonetically here: "car-wreck") for this package, with an additional shorthand of carr that does the same thing.

The goal here is to minimize the number of function clashes with multiple packages that I use in my workflow. For example: car, dplyr, and Hmisc all have recode() functions. I rely on the car package just for this function, but it conflicts with some other tidyverse functions that are vital to my workflow.

Usage

carrec(var, recodes, as_fac, as_num = TRUE, levels)

carr(...)

Arguments

Details

Recode specifications appear in a character string, separated by semicolons (see the examples below), of the form input=output. If an input value satisfies more than one specification, then the first (from left to right) applies. If no specification is satisfied, then the input value is carried over to the result. NA is allowed on input and output.

Value

carrec() returns a vector, recoded to the specifications of the user. carr() is a simple shortcut forcarrec().

Author(s)

John Fox

References

Fox, J. and Weisberg, S. (2019). An R Companion to Applied Regression, Third Edition, Sage.

Examples

x <- seq(1,10)
carrec(x,"0=0;1:2=1;3:5=2;6:10=3")

Description

Scoped helper verbs included in this R Documentation file allow for targeted commands on specified columns. They also rename the ensuing output to conform to my preferred style. The commands here are multiple and explained in the details section below.

Usage

center_at(data, x, prefix = "c", na = TRUE, .by = NULL)

diff_at(data, x, o = 1, prefix = "d", .by = NULL)

group_mean_center_at(
  data,
  x,
  mean_prefix = "mean",
  prefix = "b",
  na = TRUE,
  .by
)

lag_at(data, x, prefix = "l", o = 1, .by = NULL)

log_at(data, x, prefix = "ln", plus_1 = FALSE)

mean_at(data, x, prefix = "mean", na = TRUE, .by = NULL)

r1sd_at(data, x, prefix = "s", na = TRUE, .by = NULL)

r2sd_at(data, x, prefix = "z", na = TRUE, .by = NULL)

Arguments

Details

center_at is a wrapper for mutate_at and rename_at from dplyr. It takes supplied vectors and effectively centers them from the mean. It then renames these new variables to have a prefix of c_. The default prefix ("c") can be changed by way of an argument in the function.

diff_at is a wrapper for mutate and across from dplyr. It takes supplied vectors and creates differences from the previous value recorded above it. It then renames these new variables to have a prefix of d_ (in the case of a first difference), or something like d2_ in the case of second differences, or d3_ in the case of third differences (and so on). The exact prefix depends on the o argument, which communicates the order of lags you want. It defaults to 1. The default prefix ("d") can be changed by way of an argument in the function, though the naming convention will omit a numerical prefix for first differences.

group_mean_center_at is a wrapper for mutate and across in dplyr. It takes supplied vectors and centers an (assumed) group mean of the variables from an (assumed) total population mean of the variables provided to it. It then returns the new variables with a prefix, whose default is b_. This prefix communicates, if you will, a kind of "between" variable in the panel model context, in juxtaposition to "within" variables in the panel model context.

lag_at is a wrapper for mutate and across from dplyr. It takes supplied vector(s) and creates lag variables from them. These new variables have a prefix of l[o]_ where o corresponds to the order of the lag (specified by an argument in the function, which defaults to 1). This default prefix ("l") can be changed by way of an another argument in the function.

log_at is a wrapper for mutate and across from dplyr. It takes supplied vectors and creates a variable that takes a natural logarithmic transformation of them. It then renames these new variables to have a prefix of ln_. This default prefix ("ln") can be changed by way of an argument in the function. Users can optionally specify that they want to add 1 to the vector before taking its natural logarithm, which is a popular thing to do when positive reals have naturally occurring zeroes.

mean_at is a wrapper for mutate and across from dplyr. It takes supplied vectors and creates a variable communicating the mean of the variable. It then renames these new variables to have a prefix of mean_. This default prefix ("mean") can be changed by way of an argument in the function.

r1sd_at is a wrapper for mutate and across from dplyr. It both rescales the supplied vectors to new vectors and renames the vectors to each have a prefix of s_. Note the rescaling here is just by one standard deviation and not two. The default prefix ("s") can be changed by way of an argument in the function.

r2sd_at is a wrapper for mutate and across from dplyr. It both rescales the supplied vectors to new vectors and renames the vectors to each have a prefix of z_. Note the rescaling here is by two standard deviations and not one. The default prefix ("z") can be changed by way of an argument in the function.

All functions, except for lag_at, will fail in the absence of a character vector of a length of one. They are intended to work across multiple columns instead of just one. If you are wanting to create one new variable, you should think about using some other dplyr verb on its own.

Value

The function returns a set of new vectors in a data frame after performing relevant functions. The new vectors have distinct prefixes corresponding with the action performed on them.

Examples

set.seed(8675309)
Example <- data.frame(category = c(rep("A", 5),
                                   rep("B", 5),
                                   rep("C", 5)),
                      x = runif(15), y = runif(15),
                      z = sample(1:20, 15, replace=TRUE))

my_vars <- c("x", "y", "z")
center_at(Example, my_vars)

diff_at(Example, my_vars)

diff_at(Example, my_vars, o=3)

lag_at(Example, my_vars)

lag_at(Example, my_vars, o=3)

log_at(Example, my_vars)

log_at(Example, my_vars, plus_1 = TRUE)

mean_at(Example, my_vars)

r1sd_at(Example, my_vars)

r2sd_at(Example, my_vars)

Description

This is a toy panel data set on charitable contributions across 10 years for 47 taxpayers. It's useful for illustrating the estimation of panel models.

Usage

charitable_contributions

Format

A data frame with 470 observations on the following 8 variables.

subject

a numeric identifier for the subject

time

a numeric time identifier, as a simple integer from 1 to 10

charity

the sum of cash and other property contributions, excluding carry-overs from previous years

income

adjusted gross income

price

1 minus the marginal income tax rate, which is defined on income prior to contributions

age

a dummy variable that equals 1 if the respondent is over 64, 0 otherwise

ms

a dummy variable that equals 1 if the respondent is married, 0 otherwise

deps

the number of claimed dependents, as an integer

Details

Frees (2003) is the nominal source for these data, as they appear as toy data sets for use in his book. He in turn cites Banerjee and Frees (1995), though this citation may have been meant for a 1997 article in Journal of the American Statistical Association. The actual source for these data as I obtained them is Gujarati (2012). The underlying source of the raw data are supposedly the 1979-1988 Statistics of Income Panel of Individual Tax Returns. Given the opacity of the data, and its temporal limitations, these data should only be used for illustration and not inference.

The charitable contributions variable and income variables are very clearly log-transformed. Banerjee and Price (1997) seem to imply the price variable is as well.

Description

A function to simulate data from a correlation matrix. This is useful for illustrating some theoretical properties of regressions when population parameters are known and set in advance.

Usage

cor2data(cor, n, seed)

Arguments

Value

cor2data() returns a data frame where all observations are simulated from a standard normal distribution, but with those pre-set correlations.

Author(s)

Steven V. Miller

Examples

vars <- c("control", "treat", "instr", "e")
Correlations <- matrix(cbind(1, 0.001, 0.001, 0.001,
                             0.001, 1, 0.85, -0.5,
                            0.001, 0.85, 1, 0.001,
                           0.001, -0.5, 0.001, 1),nrow=4)

rownames(Correlations) <- colnames(Correlations) <- vars

cor2data(Correlations, 1000, 8675309)

Description

corvectors() is a function to obtain a multivariate dataset by specifying the relation between those specified variables.

Usage

corvectors(
  data,
  corm,
  tol = 0.005,
  conv = 10000,
  cores = 2,
  splitsize = 1000,
  verbose = FALSE,
  seed
)

Arguments

Details

This is liberally copy-pasted from van Kooten and Vink's wonderful-but-no-longer-supported correlate package. They call it correlate() in their package, but I opt for corvectors() here.

Value

corvectors() returns a matrix given the specified multivariate relation.

Author(s)

Pascal van Kooten and Gerko Vink

Examples

## Not run: 
set.seed(8675309)
library(tibble)
# bivariate example, start with zero correlation
as_tibble(data.frame(corvectors(replicate(2, rnorm(100)), .5)))

# multivariate example
as_tibble(data.frame(corvectors(replicate(4, rnorm(100)), c(.5, .6, .7))))


## End(Not run)

Description

db_lselect() allows you to select variables from multiple tables in an SQL database. It returns a lazy query that combines all the variables together into one data frame (as a tibble). The user can choose to run collect() after this query if they see fit.

Usage

db_lselect(.data, connection, vars)

Arguments

Details

This is a wrapper function in which purrr and dplyr are doing the heavy lifting. The tables in the database are declared as a character (or character vector). The variables to select are also declared as a character (or character vector), which are then wrapped in a one_of() function within select() in dplyr.

Value

Assuming a particular structure to the database, the function returns a combined table including all the requested variables from all the tables listed in the data character vector. The returned table will have other attributes inherited from how dplyr interfaces with SQL, allowing the user to extract some information about the query (e.g. through show_query()).

References

Miller, Steven V. 2020. "Clever Uses of Relational (SQL) Databases to Store Your Wider Data (with Some Assistance from dplyr and purrr)" http://svmiller.com/blog/2020/11/smarter-ways-to-store-your-wide-data-with-sql-magic-purrr/

Examples

library(DBI)
library(RSQLite)
library(dplyr)
library(dbplyr)
set.seed(8675309)

A <- data.frame(uid = c(1:10),
                a = rnorm(10),
                b = sample(letters, 10),
                c = rbinom(10, 1, .5))

B <- data.frame(uid = c(11:20),
                a = rnorm(10),
                b = sample(letters, 10),
                c = rbinom(10, 1, .5))

C <- data.frame(uid = c(21:30), a = rnorm(10),
                b = sample(letters, 10),
                c = rbinom(10, 1, .5),
                d = rnorm(10))

con <- dbConnect(SQLite(), ":memory:")

copy_to(con, A, "A",
        temporary=FALSE)

copy_to(con, B, "B",
        temporary=FALSE)

copy_to(con, C, "C",
        temporary=FALSE)

# This returns no warning because columns "a" and "b" are in all tables
c("A", "B", "C") %>% db_lselect(con, c("uid", "a", "b"))

# This returns two warnings because column "d" is not in 2 of 3 tables.
# ^ this is by design. It'll inform the user about data availability.
c("A", "B", "C") %>% db_lselect(con, c("uid", "a", "b", "d"))
dbDisconnect(con)

Description

These are data to illustrate labeled data and how to process them with get_var_info() in this package.

Usage

ess9_labelled

Format

A data frame with 109 observations on the following 4 variables.

essround

a numeric constant

edition

another numeric constant

cntry

a character vector (with label) for the country in the data

netusoft

a numeric vector (with label) for self-reported internet consumption of a respondent

Details

Data are condensed summaries from the raw data. They amount to every unique combination of country and self-reported internet consumption. The data are here to illustrate the get_var_info() function in this package.

Description

fct_reorg() is a forcats hack that reorganizes a factor after re-leveling it. It has been situationally useful in my coefficient plots over the years.

Usage

fct_reorg(fac, ...)

Arguments

Details

Solution comes by way of this issue on Github: https://github.com/tidyverse/forcats/issues/45

Value

This function takes a character or factor vector and first re-levels it before re-coding certain values. The end result is a factor.

Examples

x<-factor(c("a","b","c"))
fct_reorg(x, B="b", C="c")

Description

filter_refs() is a convenience function I wrote for filtering a data frame of citations returning the entries as a valid .bib entry (as a character vector). I wrote this for more easily passing on citations to the print_refs() function also included in this package.

Usage

filter_refs(bibdat, criteria, type = "bibtexkey")

Arguments

Details

filter_refs() assumes some familiarity with BibTeX, .bib entries, and depends on the bib2df package.

Value

filter_refs() takes a data frame of citations, like the one created by the bib2df package, and returns a character vector (amounting to a valid .bib entry) of citations the user wants. This can then be easily passed to the print_refs() function also included in this package.

Examples

# Based on `stevepubs` configuration, filter on `BIBTEXKEY` where
# the citation key matches one of these.
filter_refs(stevepubs, c("miller2017etst", "miller2017etjc", "miller2013tdpi"))

# Based on `stevepubs` configuration, filter on `YEAR` where
# the publication year is 2017, 2018, 2019, 2020, or 2021.
filter_refs(stevepubs, c(2017:2021), type = "year")

Description

These are data generated in peacesciencer for all French leader-years from 1874 to 2015. I'm going to use these data for stress-testing the calculation of so-called "peace spells" for data that are decidedly imbalanced, as these are.

Usage

fra_leaderyears

Format

A data frame with 255 observations on the following 10 variables.

obsid

the unique observation ID in the Archigos data

ccode

the Correlates of War state code for France (220)

leader

a name—typically last name—for the leader

year

an observation year for the leader

startdate

the start date for the leader's period in office

enddate

the end date for the leader's period in office

gmlmidongoing

was there an ongoing inter-state dispute for the leader?

gmlmidonset

was there a new inter-state dispute onset for the leader?

gmlmidongoing_init

was there an ongoing inter-state dispute for the leader that the leader initiated?

gmlmidonset_init

was there a new inter-state dispute onset for the leader that the leader initiated?

Details

Data are generated in the development version (scheduled release of v. 0.7) of peacesciencer. Conflict data come from the GML MID data (v. 2.2.1). Leader data come from Archigos (v. 4.1).

References

Goemans, Henk E., Kristian Skrede Gleditsch, and Giacomo Chiozza. 2009. "Introducing Archigos: A Dataset of Political Leaders" Journal of Peace Research 46(2): 269–83.

Gibler, Douglas M., Steven V. Miller, and Erin K. Little. 2016. “An Analysis of the Militarized Interstate Dispute (MID) Dataset, 1816-2001.” International Studies Quarterly 60(4): 719-730.

Description

get_sims() is a function to simulate quantities of interest from a multivariate normal distribution for "new data" from a regression model.

Usage

get_sims(model, newdata, nsim, seed)

Arguments

Details

This (should) be a flexible function that takes a merMod object (estimated from lme4, blme, etc.) or a lm or glm object and generates some quantities of interest when paired with new data of observations of interest. Of note: I've really only tested this function with linear models, generalized linear models, and their mixed model equivalents. For mixed models, this approach does not offer support for the incorporation of the random effects or the random slopes. It's just for the fixed effects, which is typically what most people want anyway. Users who want to better incorporate the random intercepts or slope could find that support in the merTools package.

Value

get_sims() returns a data frame (as a tibble) with the quantities of interest and identifying information about the particular simulation number.

Author(s)

Steven V. Miller

Examples

## Not run: 
# Note: these models are dumb, but they illustrate how it works.

M1 <- lm(mpg ~ hp, mtcars)
# Note: this function requires the DV to appear somewhere, anywhere in the "new data"
newdat <- data.frame(mpg = 0,
                     hp = c(mean(mtcars$hp) - sd(mtcars$hp),
                            mean(mtcars$hp),
                            mean(mtcars$hp) + sd(mtcars$hp)))

get_sims(M1, newdat, 100, 8675309)

# Note: this is likely a dumb model, but illustrates how it works.
mtcars$mpgd <- ifelse(mtcars$mpg > 25, 1, 0)

M2 <- glm(mpgd ~ hp, mtcars, family=binomial(link="logit"))

# Again: this function requires the DV to be somewhere, anywhere in the "new data"
newdat$mpgd <- 0

# Note: the simulations are returned on their original "link". Here, that's a "logit"
# You can adjust that accordingly. `plogis(y)` will convert those to probabilities.
get_sims(M2, newdat, 100, 8675309)

library(lme4)
M3 <- lmer(mpg ~ hp + (1 | cyl), mtcars)

# Random effects are not required here since we're passing over them.
get_sims(M3, newdat, 100, 8675309)

## End(Not run)

Description

get_var_info() allows you to peek at your labelled data, extracting a given column's variable labels. The intended use here is mostly "peeking" for the purpose of recoding column's in the absence of a codebook or other form of documentation. gvi() is a shortcut for this function.

Usage

get_var_info(.data, x)

gvi(...)

Arguments

Details

This function leans on var_label() and val_label() in the labelled package, which is a dependency for this package. The function is designed to be used in a "pipe."

Value

If the column in the data frame is not labelled, the function returns a message communicating the absence of labels. If the column in the data frame is labelled, the function returns a small data frame communicating the var_label() output (var), the (often but not always) numeric "code" coinciding with with the label (code), and the "label" attached to it (label).

Examples

library(tibble)
library(dplyr)
library(magrittr)

ess9_labelled %>% get_var_info(netusoft) # works, as intended
ess9_labelled %>% get_var_info(cntry) # works, as intended
ess9_labelled %>% get_var_info(ess9round) # barks at you; data are not labelled

Description

These are data generated in peacesciencer for all German (and Prussian) dyad-years from 1816 to 2020. These are going to be useful in stress-testing what "peace spell" calculations may look like when there is a huge gap in between years. In the Correlates of War context, Germany disappears from the international system from 1945 to 1990. It'll also serve as a nice test for making sure spell calculations don't misbehave in the context of missing data. In this application, there are no data for disputes between 2011 and 2020, but the dyad-years include 2011 to 2020.

Usage

gmy_dyadyears

Format

A data frame with 11174 observations on the following 6 variables.

dyad

a unique identifier for the dyad

ccode1

the Correlates of War state code for Germany (255)

ccode2

the Correlates of War state code for the other state in the dyad

year

an observation year for the dyad

gmlmidongoing

was there an ongoing inter-state dispute in the dyad-year?

gmlmidonset

was there a new inter-state dispute onset in the dyad-year

Details

Data are generated in the development version (scheduled release of v. 0.7) of peacesciencer. Conflict data come from the GML MID data (v. 2.2.1).

References

Gibler, Douglas M., Steven V. Miller, and Erin K. Little. 2016. “An Analysis of the Militarized Interstate Dispute (MID) Dataset, 1816-2001.” International Studies Quarterly 60(4): 719-730.

Description

jenny() sets a reproducible seed of 8675309. It is the only reproducible seed you should use.

Usage

jenny(x = 8675309)

Arguments

Details

jenny() comes with some additional perks if you have the emo package installed. The package is optional.

Value

When x is not specified or is 8675309, the function sets a reproducible seed of 8675309 and returns a nice message congratulating you for it. If x is not 8675309, the function sets no reproducible seed and gently admonishes you for wasting its time.

Examples

jenny() # will work and reward you for it
jenny(12345) # will not work and will result in a stern message

Description

linloess_plot() provides a visual diagnostic of the linearity assumption of the OLS model. Provided an OLS model fit by lm() in base R, the function extracts the model frame and creates a faceted scatterplot. For each facet, a linear smoother and LOESS smoother are estimated over the points. Users who run this function can assess just how much the linear smoother and LOESS smoother diverge. The more they diverge, the more the user can determine how much the OLS model is a good fit as specified. The plot will also point to potential outliers that may need further consideration.

Usage

linloess_plot(
  mod,
  resid = TRUE,
  smoother = "loess",
  se = TRUE,
  span = 0.75,
  ...
)

Arguments

Details

This function makes an implicit assumption that there is no variable in the regression formula with the name ".y" or ".resid".

It may be in your interest (for the sake of rudimentary diagnostic checks) to disable the standard error bands for particularly ill-fitting linear models.

Value

linloess_plot() returns a faceted scatterplot as a ggplot2 object. The linear smoother is in solid blue (with blue standard error bands) and the LOESS smoother is a dashed black line (with gray/default standard error bands). You can add cosmetic features to it after the fact. The function may spit warnings to you related to the LOESS smoother, depending your data. I think these to be fine the extent to which this is really just a visual aid and an informal diagnostic for the linearity assumption.

Author(s)

Steven V. Miller

Examples

M1 <- lm(mpg ~ ., data=mtcars)

linloess_plot(M1)
linloess_plot(M1, color="black", pch=21)

Description

make_perclab() takes a proportion, multiplies it by 100, optionally rounds it, and pastes a percentage sign next to it.

Usage

make_perclab(x, d = 2)

Arguments

Details

This function is useful if you're modeling proportions in something like a bar chart (for which proportions are more flexible) but want to label each bar as a percentage. The function here is mostly cosmetic.

Value

The function takes a proportion, multiplies it by 100, (optionally) rounds it to a set decimal point, and pastes a percentage sign next to it.

Examples

x <- runif(100)
make_perclab(x)

Description

make_scale() will rescale any vector to have a user-defined minimum and maximum.

Usage

make_scale(x, minim, maxim)

Arguments

Details

This function is useful if you wanted to do some kind of minimum-maximum rescaling of a variable on some given scale, prominently rescaling to a minimum of 0 and a maximum of 1 (thinking ahead to a regression). The function is flexible enough for any minimum or maximum.

Value

The function takes a numeric vector and returns a rescaled version of it with the observed (desired) minimum, the observed (desired) maximum, and rescaled values between both extremes.

Examples

x <- runif(100, 1, 100)
make_scale(x, 2, 5) # works
make_scale(x, 5, 2) # results in message
make_scale(x, 0, 1) # probably why you're using this.

Description

This is a simple data set that records every wrong guess for map quiz assignments I gave in my intro to IR class at Clemson University across five semesters.

Usage

map_quiz

Format

A data frame with 1772 observations on the following 8 variables.

class

an ordered factor of the semester in which the wrong guess was recorded by a student. Levels include "Spring 2018", "Fall 2018", "Spring 2019", "Fall 2019", and "Spring 2020."

students

the number of students in the class taking the map quiz.

region

the region map on which the country was located. Values include "Europe", "Africa", "Asia", "Latin America", and "MENA." "MENA" is short for "Middle East and North Africa."

country

the country I asked the student to correctly identify

guess

the country that was the actual state incorrectly guessed by the student

ccode1

the Correlates of War state code for the state I wanted the student to identify in country.

ccode2

the Correlates of War state code for the state that is the wrong guess for the state in guess

mindist

the minimum distance (in kilometers) between country and guess

Details

Students can always not make a guess and be wrong, which explains the NAs in the data. Students were given five separate numbered maps and prompted to identify 10 countries each on them. The maps never changed across five semesters, nor did the prompts. Use these data as you see fit. Obviously, FERPA considerations mean I can't share anything else of potential value here.

Description

mround() is a convenience function I wrote for my annotating bar charts that I make. Assuming a proportion variable, mround() will multiply each value by 100 and round it for presentation. By default, it rounds to two. The user can adjust this.

Usage

mround(x, d = 2)

Arguments

Details

This is a sister function of make_perclab() in the same package. This, however, won't add a percentage sign.

Value

The function takes a numeric vector, multiplies it by 100, rounds it (to two digits by default), and returns it to the user.

Examples

x <- runif(100)
mround(x)
mround(x, 2) # same as above
mround(x, 3)

Description

normal_dist() is a convenience function for making a plot of a normal distribution with annotated areas underneath the normal curve.

Usage

normal_dist(curvecolor, fillcolor, fontfamily)

Arguments

Details

The normal distribution is a standard normal distribution with a mean of 0 and a standard deviation of 1.

Value

The function returns a fancy plot of a normal distribution annotated with areas underneath the hood. Note that whatever color is supplied in fillcolor is automatically lightened for areas further from the center of the curve.

Examples

library(stevemisc)
normal_dist("blue","red")
normal_dist("purple","orange")

Description

I loathe how statistical instruction privileges obtaining a magical p-value by reference to an area underneath the standard normal curve, only to botch what the actual z-value is corresponding to the magical p-value. This simple function converts the p-value you want (typically .05, thanks to R.A. Fisher) to the z-value it actually is for the kind of claims we typically make in inferential statistics. If we're going to do inference the wrong way, let's at least get the z-value right.

Usage

p_z(x, ts = TRUE)

Arguments

Details

p_z() takes a p-value of interest and converts it, with precision, to the z-value it actually is. The function takes a vector and returns a vector. The function assumes you're doing something akin to calculating a confidence interval or testing a regression coefficient against a null hypothesis of zero. This means the default output is a two-sided critical z-value. We're taught to use two-sided z-values when we're agnostic about the direction of the effect or statistic of interest, which is, to be frank, hilarious given how most research is typically done.

Value

This function takes a numeric vector, corresponding to the p-value you want, and returns a numeric vector coinciding with the z-value you want under the standard normal distribution. For example, the z-value corresponding with the magic number of .05 (the conventional cutoff for assessing statistical significance) is not 1.96, it's something like 1.959964 (rounding to the default six decimal points).

Examples

library(stevemisc)

p_z(.05)
p_z(c(.001, .01, .05, .1))
p_z(.05, ts=FALSE)
p_z(c(.001, .01, .05, .1), ts=FALSE)

Description

prepare_refs does some last-minute formatting of a data frame created by bib2df so that it can be formatted nicely to various outputs.

Usage

prepare_refs(bib2df_refs, toformat = "plain")

Arguments

Details

The function is designed to work more generally in the absence of various fields. Assume, for example, that your data frame has no BOOK field. The function uses the one_of() wrapper to work around this. The "warning" returned by the function is more of a message. This function may be expanded as I think of more use cases.

Value

print_refs() does some last-minute formatting to a data frame created by bib2df so that rendering in R Markdown is a little easier and less code-heavy.

See Also

print_refs() for formatting a .bib references to various outputs.

Examples

prepare_refs(stevepubs)

Description

print_refs() is a convenience function I found and edited that will allow a user to print and format .bib entries as if they were references. This function is useful if you want to load a .bib entry or set of entries and print them in the middle of a document in R Markdown.

Usage

print_refs(
  bib,
  csl = "american-political-science-association.csl",
  toformat = "markdown_strict",
  cslrepo = "https://raw.githubusercontent.com/citation-style-language/styles/master",
  spit_out = TRUE,
  delete_after = TRUE
)

Arguments

Details

print_refs() assumes an active internet connection in the absence of the appropriate CSL file in the working directory. The citation style language (CSL) file supplied by the user must match a file in the massive Github repository of CSL files. Users interested in potential outputs should read more about Pandoc (https://pandoc.org/MANUAL.html). The Github repository of CSL files is available here: https://github.com/citation-style-language/styles.

Value

print_refs() takes a .bib entry and returns the requested formatted reference or references from it.

Examples

example <- "@Book{vasquez2009twp, Title = {The War Puzzle Revisited},
Author = {Vasquez, John A}, Publisher = {New York, NY: Cambridge University Press},
Year = {2009}}"

print_refs(example)

Description

ps_btscs() allows you to create spells ("peace years" in the international conflict context) between observations of some event. This will allow the researcher to better model temporal dependence in binary time-series cross-section ("BTSCS") models. It is an improvement on sbtscs() (included in this package) by its ability to more flexibly work with data that have lots of NAs that bracket the observed event data. It is used in the peacesciencer package.

Usage

ps_btscs(data, event, tvar, csunit, pad_ts = FALSE)

Arguments

Details

This function is derived from sbtscs(). See documentation there for more information.

Value

ps_btscs() takes a data frame and returns the data frame with a new variable named spell.

Author(s)

David A. Armstrong, Steven V. Miller

References

Armstrong, Dave. 2016. “DAMisc: Dave Armstrong's Miscellaneous Functions.” R package version 1.4-3.

Miller, Steven V. 2017. “Quickly Create Peace Years for BTSCS Models with sbtscs in stevemisc.” http://svmiller.com/blog/2017/06/quickly-create-peace-years-for-btscs-models-with-stevemisc/

Examples

library(dplyr)
library(stevemisc)
data(usa_mids)

# notice: no quotes
ps_btscs(usa_mids, midongoing, year, dyad)

Description

ps_spells() allows you to create spells ("peace years" in the international conflict context) between observations of some event. This will allow the researcher to better model temporal dependence in binary time-series cross-section ("BTSCS") models. The function is one of three in this package, and the contents of this function are partly ported from the add_duration() function in the spduration package. That function, unlike the other two I offer here, works much better where panels are decidedly imbalanced.

Usage

ps_spells(data, event, tvar, csunit, time_type = "year", ongoing = FALSE)

Arguments

Details

This function is derived from add_duration() in the spduration package. See documentation there for more information. I thank Andreas Beger for the blessing to port parts of it here.

Value

ps_spells() takes a data frame and returns the data frame with a new variable named spell.

Author(s)

Andreas Beger, Steven V. Miller

References

Beger, Andreas, Daina Chiba, Daniel W. Hill, Jr, Nils W. Metternich, Shahryar Minhas and Michael D. Ward. 2018. “spduration: Split-Population and Duration (Cure) Regression.” R package version 0.17.1.

Examples

One <- ps_btscs(usa_mids, midongoing, year, dyad)
Two <- ps_spells(usa_mids, midongoing, year, dyad)
identical(One, Two)

Description

r1sd allows you to rescale a numeric vector such that the ensuing output has a mean of 0 and a standard deviation of 1.

Usage

r1sd(x, na = TRUE)

Arguments

Details

This is a convenience function since the default rescale() function has some additional weirdness that is not welcome for my use cases. By default, na.rm is set to TRUE.

Value

The function returns a numeric vector rescaled with a mean of 0 and a standard deviation of 1.

Examples

x <- rnorm(100)
r1sd(x)

Description

r2sd allows you to rescale a numeric vector such that the ensuing output has a mean of 0 and a standard deviation of .5. r2sd_at is a wrapper for mutate_at and rename_at from dplyr. It both rescales the supplied vectors to new vectors and renames the vectors to each have a prefix of z_.

Usage

r2sd(x, na = TRUE)

Arguments

Details

By default, na.rm is set to TRUE. If you have missing data, the function will just pass over them.

Gelman (2008) argues that rescaling by two standard deviations puts regression inputs on roughly the same scale no matter their original scale. This allows for some honest, if preliminary, assessment of relative effect sizes from the regression output. This does that, but without requiring the rescale function from arm. I'm trying to reduce the packages on which my workflow relies.

Importantly, I tend to rescale only the ordinal and interval inputs and leave the binary inputs as 0/1. So, my r2sd function doesn't have any of the fancier if-else statements that Gelman's rescale function has.

Value

The function returns a numeric vector rescaled with a mean of 0 and a standard deviation of .5.

References

Gelman, Andrew. 2008. "Scaling Regression Inputs by Dividing by Two Standard Deviations." Statistics in Medicine 27: 2865–2873.

Examples

x <- rnorm(100)
r2sd(x)

r2sd_at(mtcars, c("mpg", "hp", "disp"))

Description

rbnorm() is a function to randomly generate values from a bounded normal (really: a scaled beta) distribution with specified mean, standard deviation, and upper/lower bounds. I use this function to randomly generate data that we treat as interval for sake of getting means and standard deviations, but have discernible bounds (and even skew) to teach students about things like random sampling and central limit theorem.

Usage

rbnorm(n, mean, sd, lowerbound, upperbound, round = FALSE, seed)

Arguments

Details

I call it "bounded normal" when it's really a beta distribution. I'm aware of this. I took much of this code from somewhere. I forget where.

Value

The function returns a vector of simulated data approximating the user-specified conditions.

Examples

library(tibble)

tibble(x = rbnorm(10000, 57, 14, 0, 100))
tibble(x = rbnorm(10000, 57, 14, 0, 100, round = TRUE))
tibble(x = rbnorm(10000, 57, 14, 0, 100, seed = 8675309))

Description

rd_plot() provides a visual diagnostic of the normality assumption of the linear model. Provided an OLS model fit by lm() in base R, the function extracts the residuals of the model and creates a density plot of those residuals (solid black line) against a standard normal distribution with a mean of 0 and a standard deviation matching the standard deviation of the residuals from the model. The function may be used for diagnostic purposes.

Usage

rd_plot(mod)

Arguments

Details

The user can always add ggplot2 elements on top of this for greater legibility/clarity. For example, density plots can be finicky about making observations appear where they don't. Perhaps adjusting the scale of x ad hoc, after the fact, may be warranted.

The goal of this function is to emphasize that in many real world applications, the normality assumption of the residuals is never held but can often be reasonably approximated upon visual inspection.

Value

rd_plot() returns a density plot a ggplot2 object. A density plot of the actual residuals is a solid black line. A stylized normal distribution matching the description of the residuals is the blue dashed line.

Author(s)

Steven V. Miller

Examples

M1 <- lm(mpg ~ ., data=mtcars)

rd_plot(M1)

Description

revcode allows you to reverse code a numeric variable. If, say, you have a Likert item that has values of 1, 2, 3, 4, and 5, the function inverts the scale so that 1 = 5, 2 = 4, 3 = 3, 4 = 2, and 5 = 1.

Usage

revcode(x)

Arguments

Details

This function passes over NAs you may have in your variable. It does assume, reasonably might I add, that the observed values include both the minimum and the maximum. This is usually the case in a discrete ordered-categorical variable (like a Likert item). It also assumes that the numeric vector supplied to it contains all possible values and that the minimum observed value is 1. This is usually a safe assumption in survey data where the variable of interest is ordinal (either on a 1:4 scale, or 1:5 scale, or 1:10 scale). No matter, use the function with that in mind.

Value

The function returns a numeric vector that reverse codes the the numeric vector that was supplied to it.

Examples

data.frame(x1 = rep(c(1:7, NA), 2),
      x2 = c(1:10, 1:4, NA, NA),
     x3 = rep(c(1:4), 4)) -> example_data

library(dplyr)
library(magrittr)

example_data %>% mutate_at(vars("x1", "x2", "x3"), ~revcode(.))

Description

sbayesboot() performs a Bayesian bootstrap of a regression model.

Usage

sbayesboot(object, reps = 1000L, seed, cluster = NULL, ...)

Arguments

Details

The code underpinning sbayesboot() is largely derived from code provided by Grant McDermott and Vincent Arel-Bundock. My approach here takes the flexibility of McDermott's model-agnostic code (along with the ease of specifying clusters) and combines it with Arel-Bundock's update() approach to the actual bootstrapping. I may have screwed something up, so feel free to point to cases where I did screw up.

Value

sbayesboot() takes a fitted regression model and returns a matrix of bootstrapped coefficients (with intercept). These could be easily converted to a data frame for ease of summary.

Author(s)

Grant McDermott, Vincent Arel-Bundock

Examples

M1 <- lm(mpg ~ disp + wt + hp, mtcars)

# Default options

BB1 <- sbayesboot(M1)

# Cluster bootstrap on cylinder variable

BB2 <- sbayesboot(M1, cluster=~cyl)

Description

sbtscs() allows you to create spells ("peace years" in the international conflict context) between observations of some event. This will allow the researcher to better model temporal dependence in binary time-series cross-section ("BTSCS") models.

Usage

sbtscs(data, event, tvar, csunit, pad_ts = FALSE)

Arguments

Details

I should confess outright, and it should be obvious to anyone who looks at the code, that I liberally copy from Dave Armstrong's btscs() function in the DAMisc package. I offer two such improvements. One, the btscs() function chokes when a large number of cross-sectional units have no recorded "event." I don't know why this happens but it does. Further, "tidying" up the code by leaning on dplyr substantially speeds up computation. Incidentally, this concerns the same cross-sectional units with no recorded events that can choke the btscs() function in large numbers.

Value

sbtscs() takes a data frame and returns the data frame with a new variable named spell.

Author(s)

David A. Armstrong, Steven V. Miller

References

Armstrong, Dave. 2016. “DAMisc: Dave Armstrong's Miscellaneous Functions.” R package version 1.4-3.

Miller, Steven V. 2017. “Quickly Create Peace Years for BTSCS Models with sbtscs in stevemisc.” http://svmiller.com/blog/2017/06/quickly-create-peace-years-for-btscs-models-with-stevemisc/

Examples

## Not run: 
library(dplyr)
library(stevemisc)
data(usa_mids)

# notice: no quotes
sbtscs(usa_mids, midongoing, year, dyad)

## End(Not run)

Description

show_ranef() allows a user estimating a mixed model to quickly plot the random intercepts (with conditional variances) of a given random effect in a mixed model. In cases where there is a random slope over the intercept, the function plots the random slope as another caterpillar plot (as another facet)

Usage

show_ranef(model, group, reorder = TRUE)

Arguments

Details

This function is a simple wrapper in which broom.mixed and, obviously ggplot2 are doing the heavy lifting.

Value

show_ranef() returns a caterpillar plot of the random intercepts from a given mixed model. If broom.mixed::augment() can process it, this function should work just fine.

Author(s)

Steven V. Miller

Examples

library(lme4)
library(stevemisc)
data(sleepstudy)

M1 <- lmer(Reaction ~ Days + (Days | Subject), data=sleepstudy)
show_ranef(M1, "Subject")
show_ranef(M1, "Subject", reorder=FALSE)

Description

smvrnorm() simulates data from a multivariate normal distribution.

Usage

smvrnorm(
  n = 1,
  mu,
  sigma,
  tol = 1e-06,
  empirical = FALSE,
  eispack = FALSE,
  seed
)

Arguments

Details

This is a simple port and rename of mvrnorm() from the MASS package. I elect to plagiarize/port it because the MASS package conflicts with a lot of things in my workflow, especially select(). This is useful for "informal Bayes" approaches to generating quantities of interest from a regression model.

Value

The function returns simulated data from a multivariate normal distribution.

References

B. D. Ripley (1987) Stochastic Simulation. Wiley. Page 98.

Examples

M1 <- lm(mpg ~ disp + cyl, mtcars)

smvrnorm(100, coef(M1), vcov(M1))

Description

These are data on my publications, barring a few things like book reviews and some forthcoming pieces. I use these data to illustrate the print_refs() function. You should cite my publications more.

Usage

stevepubs

Format

A data frame with the following 14 variables.

CATEGORY

the entry type

BIBTEXKEY

the unique entry key

AUTHOR

a list of authors for this entry

BOOKTITLE

the book title, if appropriate

JOURNAL

the journal title, if appropriate

NUMBER

the journal volume number, if appropriate

PAGES

the range of page numbers, if appropriate

PUBLISHER

the book publisher, if appropriate

TITLE

the title of the publication

VOLUME

the journal volume number, if appropriate

YEAR

the year of publication, as a character. Publications with no year are assumed to be forthcoming

DOI

a DOI, if I entered one

Details

Cite my publications more, you goons. Extremely Smokey Bear voice Only YOU can jack my h-index to infinity.

Description

A simple summary of all strategic (inter-state) rivalries from Thompson and Dreyer (2012).

Usage

data("strategic_rivalries")

Format

A data frame with 197 observations on the following 10 variables.

rivalryno

a numeric vector for the rivalry number

rivalryname

a character vector for the rivalry name

sidea

a character vector for the first country in the rivalry

sideb

a character vector for the second country in the rivalry

styear

a numeric vector for the start year of the rivalry

endyear

a numeric vector for the end year of the rivalry

region

a character vector for the region of the rivalry, per Thompson and Dreyer (2012)

type1

a character vector for the primary type of the rivalry (spatial, positional, ideological, or interventionary)

type2

a character vector for the secondary type of the rivalry, if applicable (spatial, positional, ideological, or interventionary)

type3

a character vector for the tertiary type of the rivalry, if applicable (spatial, positional, ideological, or interventionary)

Details

Information gathered from the appendix of Thompson and Dreyer (2012). Ongoing rivalries are right-bound at 2010, the date of publication for Thompson and Dreyer's handbook. Users are free to change this if they like.

References

Thompson, William R. and David Dreyer. 2012. Handbook of International Rivalries. CQ Press.

Examples

data(strategic_rivalries)

Description

These are density, distribution function, quantile function and random generation for the Student-t distribution with location mu, scale sigma, and degrees of freedom df. Base R gives you the so-called "standard" Student-t distribution, with just the varying degrees of freedom. This generalizes that standard Student-t to the three-parameter version.

Usage

dst(x, df, mu, sigma)

pst(q, df, mu, sigma)

qst(p, df, mu, sigma)

rst(n, df, mu, sigma)

Arguments

Details

This is a simple hack taken from Wikipedia. It's an itch I've been wanting to scratch for a while. I can probably generalize this outward to allow the tail and log stuff, but I wrote this mostly for the random number generation. Right now, I haven't written this to account for the fact that sigma should be non-negative, but that's on the user to know that (for now).

Value

dst() returns the density. pst() returns the distribution function. qst() returns the quantile function. rst() returns random numbers.

See Also

TDist

Description

tbl_df() ensures legacy compatibility with some of my scripts since the function is deprecated in dplyr. to_tbl() also added for fun.

Usage

tbl_df(...)

to_tbl(...)

Arguments

Value

This function takes a data frame and turns it into a tibble.

Examples

tbl_df(mtcars)
tbl_df(iris)

Description

theme_steve(), now in stevethemes, was a preferred theme of mine a few years ago. It was basically theme_bw() from ggplot2 theme, but with me tweaking a few things. I then moved to theme_steve_web() for most things now, prominently on my website. This theme incorporates the "Open Sans" and "Titillium Web" fonts that I like so much. post_bg() is a legacy function for changing the backgrounds on plots to better match what was the background color on my website. theme_steve_ms() is for LaTeX manuscripts that use the cochineal font package. theme_steve_font() is for any purpose, allowing you to supply your own font.

Usage

theme_steve_web(...)

post_bg(...)

theme_steve_ms(axis_face = "italic", caption_face = "italic", ...)

theme_steve_font(axis_face = "italic", caption_face = "italic", font, ...)

Arguments

Details

theme_steve_web() and theme_steve_ms() both explicitly depend on having the fonts installed on your end. It's ultimately optional for you to have them but the use of these functions imply them. All functions that remain here should be understood as "legacy" functions that will no longer be maintained or updated. The stevethemes package will have all my ggplot2 elements going forward.

Value

post_bg() takes a ggplot2 plot and changes the background to have a color of "#fdfdfd". theme_steve_web() extends theme_steve() to add custom fonts, notably "Open Sans" and "Titillium Web". In all cases, these functions take a ggplot2 plot and return another ggplot2 plot, but with some cosmetic changes. theme_steve_ms() takes a ggplot2 plot and overlays "Crimson Pro" fonts, which is the basis of the cochineal font package in LaTeX. theme_steve_font() takes a ggplot2 plot and overlays a font of your choosing.

See Also

ggplot2::theme

Examples

## Not run: 
library(ggplot2)


ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point() + theme_steve_web() +
  labs(title = "A ggplot2 Plot from the Motor Trend Car Road Tests Data",
  subtitle = "Notice the prettier fonts, if you have them.",
  caption = "Data: ?mtcars in {datasets} in base R.")

ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point() + theme_steve_web() +
  post_bg() +
  labs(title = "A ggplot2 Plot from the Motor Trend Car Road Tests Data",
  subtitle = "Notice the slight change in background color",
  caption = "Data: ?mtcars in {datasets} in base R.")

ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point() + theme_steve_ms() +
  labs(title = "A ggplot2 Plot from the Motor Trend Car Road Tests Data",
  subtitle = "Notice the fonts will match the 'cochineal' font package in LaTeX.",
  caption = "Data: ?mtcars in {datasets} in base R.")

ggplot(mtcars, aes(x = mpg, y = hp)) +
  geom_point() + theme_steve_font(font = "Comic Sans MS") +
  labs(title = "A ggplot2 Plot from the Motor Trend Car Road Tests Data",
  subtitle = "Notice that this will look ridiculous",
  caption = "Data: ?mtcars in {datasets} in base R.")

## End(Not run)

Description

This is a non-directed dyad-year data set for militarized interstate disputes involving the United States. I created these to illustrate the sbtscs() function.

Usage

usa_mids

Format

A data frame with 14586 observations on the following 6 variables.

dyad

a unique identifier for the dyad

ccode1

the Correlates of War state code for the United States (2)

ccode2

the Correlates of War state code for the other state in the dyad

year

an observation year for the dyad

midongoing

was there an ongoing inter-state dispute in the dyad-year?

midonset

was there a new inter-state dispute onset in the dyad-year

Details

Data were generated some time ago. Rare cases where there were multiple disputes ongoing in a given dyad-year were first whittled by isolating 1) unique dispute onsets. Thereafter, the data select the 2) highest fatality, then 3) the highest hostility level, and then 4) the longer dispute, until 5) just picking whichever one came first. There are no duplicate non-directed dyad-year observations.

References

Gibler, Douglas M., Steven V. Miller, and Erin K. Little. 2016. “An Analysis of the Militarized Interstate Dispute (MID) Dataset, 1816-2001.” International Studies Quarterly 60(4): 719-730.

Description

wls() takes an OLS model and re-estimates it using a weighted least squares approach. Weighted least squares is often a "textbook" approach to dealing with the presence of heteroskedastic standard errors, for which the weighted least squares estimates are compared to the OLS estimates of uncertainty to check for consistency or potential inferential implications.

Usage

wls(mod)

Arguments

Details

The function should be robust to potential model specification oddities (e.g. polynomials and fixed effects). It also should perform nicely in the presence of missing data, if and only if na.action = na.exclude is supplied first to the offending OLS model supplied to the function for a weighted least squares re-estimation.

Value

wls() returns a new model object that is a weighted least squares re-estimation of the OLS model supplied to it.

Author(s)

Steven V. Miller

Examples

M1 <- lm(mpg ~ ., data=mtcars)
M2 <- wls(M1)

summary(M2)

Description

wom() is a convenience function I use for constructing calendars in ggplot2. It takes a date and returns, as a numeric vector, the week of the month for the date given to it.

Usage

wom(x)

Arguments

Details

wom() assumes Sunday is the start of the week. This can assuredly be customized later in this function, but right now the assumption is Sunday is the start of the week (and not Monday, as it might be in other contexts).

Value

wom() is a convenience function I use for constructing calendars in ggplot2. It takes a date and returns, as a numeric vector, the week of the month for the date given to it.

Examples

wom(as.Date("2022-01-01"))

wom(Sys.Date())