Using the functions of the ExPanDaR package
Joachim Gassen
- -2018-05-11
- - + +2018-10-12
+ +ExPanDaR-functions.RmdWhile the main purpose of the ExPanDaR package is to provide the infrastructure for the ExPanD app, the auxiliary functions of the package can also be used to support your exploratory data analysis workflow in your own code. All functions are relatively thin wrappers around established R packages for graphics and HTML table presentation (ggplot2, kableExtra, stargazer). While I developed them to support analysis steps that are common with empirical archival research projects in the area of accounting and finance (which happens to be my field), I hope that they are generally useful for exploratory data analysis.
To see what ExPanDaR has to offer, let’s take a quick tour. For more detailed guidance on how to use a specific function presented below, take a look at the respective function’s help page.
Data Preparation
ExPanDaR is designed for exploratory panel data analysis (hence the name). Thus, while you can also use some functions on cross-sectional data, for most functions you will need a data frame containing your panel data. ExPanDaR expects the data to be organized in long format. This implies that each observation (a row) is identified by cross-sectional and time series identifiers and that variables are organized by columns. While you can have a vector of variables jointly determining the cross-section, the time-series needs to be identified by a unique variable. The ExPanDaR functions treat cross-sectional identifiers as factors and expect the time-series identifier to be coercible into an ordered factor.
For this walk-through I will use the data set russell_3000, which comes with the package. It contains some financial reporting and stock return data of Russell 3000 firms from Google Finance and Yahoo Finance and has been collected using the tidyquant package in the summer of 2017. A word of caution: While the data appears to be relatively decent quality I would advise against using this data for scientific work without verifying its integrity first. These are the variables included in the data.
-kable(data.frame(Variable=russell_3000_data_def$var_name,
- Definition=russell_3000_data_def$var_def),
+kable(data.frame(Variable=russell_3000_data_def$var_name,
+ Definition=russell_3000_data_def$var_def),
row.names = FALSE)
-
-Variable
-Definition
+
+
+Variable
+
+
+Definition
+
-
-coid
-Company identifier
+
+
+coid
+
+
+Company identifier
+
-
-period
-Fiscal year
+
+
+period
+
+
+Fiscal year
+
-
-coname
-Company name
+
+
+coname
+
+
+Company name
+
-
-sector
-Sector
+
+
+sector
+
+
+Sector
+
-
-industry
-Industry
+
+
+industry
+
+
+Industry
+
-
-toas
-Total assets at period end (M US-$)
+
+
+toas
+
+
+Total assets at period end (M US-$)
+
-
-sales
-Sales of the period (M US-$)
+
+
+sales
+
+ Sales of the period (M US-$)
+
-
-equity
-Total equity at period end (M US-$)
+
+
+equity
+
+
+Total equity at period end (M US-$)
+
-
-debt_ta
-Total debt (% total assets)
+
+
+debt_ta
+
+
+Total debt (% total assets)
+
-
-eq_ta
-Total equity (% total assets)
+
+
+eq_ta
+
+
+Total equity (% total assets)
+
-
-gw_ta
-Goodwill (% total assets)
+
+
+gw_ta
+
+
+Goodwill (% total assets)
+
-
-oint_ta
-Other intangible assets (% total assets)
+
+
+oint_ta
+
+
+Other intangible assets (% total assets)
+
-
-ppe_ta
-Property, plant and equipment (% total assets)
+
+
+ppe_ta
+
+
+Property, plant and equipment (% total assets)
+
-
-ca_ta
-Current assets (% total assets)
+
+
+ca_ta
+
+
+Current assets (% total assets)
+
-
-cash_ta
-Cash (% total assets)
+
+
+cash_ta
+
+
+Cash (% total assets)
+
-
-roe
-Return on equity (net income divided by average equity)
+
+
+roe
+
+
+Return on equity (net income divided by average equity)
+
-
-roa
-Return on assets (earnings before interest and taxes divided by average equity)
+
+
+roa
+
+
+Return on assets (earnings before interest and taxes divided by average equity)
+
-
-nioa
-Net income divided by average assets
+
+
+nioa
+
+
+Net income divided by average assets
+
-
-cfoa
-Cash flow from operations divided by average assets
+
+
+cfoa
+
+
+Cash flow from operations divided by average assets
+
-
-accoa
-Total accruals divided by average assets
+
+
+accoa
+
+
+Total accruals divided by average assets
+
-
-cogs_sales
-Cost of goods sold divided by sales
+
+
+cogs_sales
+
+
+Cost of goods sold divided by sales
+
-
-ebit_sales
-Earnings before interest and taxes divided by sales
+
+
+ebit_sales
+
+
+Earnings before interest and taxes divided by sales
+
-
-ni_sales
-Net income divided by sales
+
+
+ni_sales
+
+
+Net income divided by sales
+
-
-return
-Stock return of the period (%)
+
+
+return
+
+
+Stock return of the period (%)
+
@@ -1018,8 +1120,9 @@
Visuals
Additional visuals are available for exploring time trends. prepare_trend_graph() can be used for comparing variables…
-graph <- prepare_trend_graph(r3win[c("period", "nioa", "cfoa", "accoa")], "period")
-graph$plot
+graph <- prepare_trend_graph(r3win[c("period", "nioa", "cfoa", "accoa")], "period")
+## Warning: package 'bindrcpp' was built under R version 3.4.4
+graph$plot
… and for eyeballing the distributional properties of a single variable over time you have prepare_quantile_trend_graph().
graph <- prepare_quantile_trend_graph(r3win[c("period", "return")], "period", c(0.05, 0.25, 0.5, 0.75, 0.95))
@@ -1037,13 +1140,13 @@
Finally, if you happen to be a fan of starred numbers, you can also quickly produce regression tables by using the function prepare_regression_table() that calls lfe::felm() for OLS and glm() for binary logit models. The tables are then constructed by calling stargazer::stargazer(), allowing for plain text, html and latex output.
You can construct tables by mixing different models…
dvs <- c("return", "return", "return", "return", "return", "return")
-idvs <- list(c("nioa"),
- c("cfoa"),
- c("accoa"),
- c("cfoa", "accoa"),
- c("nioa", "accoa"),
- c("nioa", "accoa"))
-feffects <- list("period", "period", "period",
+idvs <- list(c("nioa"),
+ c("cfoa"),
+ c("accoa"),
+ c("cfoa", "accoa"),
+ c("nioa", "accoa"),
+ c("nioa", "accoa"))
+feffects <- list("period", "period", "period",
c("period", "coid"), c("period", "coid"), c("period", "coid"))
clusters <- list("", "", "", c("coid"), c("coid"), c("period", "coid"))
t <- prepare_regression_table(r3win, dvs, idvs, feffects, clusters)
@@ -1064,7 +1167,8 @@
return
return
return
-return
+return
+return
+Estimator
+ols
+ols
+ols
+ols
+ols
+ols
+
+
Fixed effects
period
period
@@ -1326,6 +1439,13 @@
+Estimator
+ols
+ols
+ols
+ols
+
+
Fixed effects
None
None
@@ -1405,7 +1525,7 @@
-
+