Codebook Demo - Freeny’s Revenue Data
Soo Chee Kiong, Ricky (17083991)
November 12th, 2020
GitHub - https://github.com/rickysoo/freeny
Contact - ricky [at] rickysoo.com
A codebook shows the structure, contents and layout of data. We should always study the codebook to understand the variables before conducting any research on the data.
The creation of codebook is demonstrated using Freeny’s Revenue Data in 2 ways - automated codebook and custom-made codebook.
Source: https://www.pexels.com
1. About The Data
The Freeny’s Revenue Data (freeny) is used in this demonstraton. The data consists of 39 observations showing quarterly revenue from 1962 2nd quarter to 1971 4th quarter.
The data originates from A. E. Freeny (1977) in A Portable Linear Regression Package with Test Programs by Bell Laboratories memorandum.
However, it remains unknown how the author created or collected the data, and whether it is based on quarterly performance on a real or fictitious company, as well as the exact units used for the measurements (thousands, millions etc.)
The data is included in the R Datasets package developed by R Core Team and contributors. The package also contains dozens of other datasets and is available in base R.
For more on the data, check out https://rdrr.io/r/datasets/freeny.html
2. Exploring Dataset
Loading Data
Load the memisc package required to create automated codebook. The package name stands for “Management of Survey Data and Presentation of Analysis Results”. The documentation can be found at https://cran.r-project.org/web/packages/memisc/memisc.pdf
library(memisc)## Loading required package: lattice## Loading required package: MASS##
## Attaching package: 'memisc'## The following objects are masked from 'package:stats':
##
## contr.sum, contr.treatment, contrasts## The following object is masked from 'package:base':
##
## as.arrayLoad the dataset.
data("freeny")Showing Data
Show the first few lines of the dataset.
head(freeny)## y lag.quarterly.revenue price.index income.level market.potential
## 1962.25 8.79236 8.79636 4.70997 5.82110 12.9699
## 1962.5 8.79137 8.79236 4.70217 5.82558 12.9733
## 1962.75 8.81486 8.79137 4.68944 5.83112 12.9774
## 1963 8.81301 8.81486 4.68558 5.84046 12.9806
## 1963.25 8.90751 8.81301 4.64019 5.85036 12.9831
## 1963.5 8.93673 8.90751 4.62553 5.86464 12.9854View the dataset in the R Studio data viewer.
View(freeny)Data Attributes
Show the dimension of the dataset. 39 observations and 5 variables are found.
dim(freeny)## [1] 39 5Show the column names. It appears that y is the revenue, and the other 4 variables are the explanatory variables.
names(freeny)## [1] "y" "lag.quarterly.revenue" "price.index"
## [4] "income.level" "market.potential"Show the row names of the dataset. It appears that each row represent a quarter of the period.
row.names(freeny)## [1] "1962.25" "1962.5" "1962.75" "1963" "1963.25" "1963.5" "1963.75"
## [8] "1964" "1964.25" "1964.5" "1964.75" "1965" "1965.25" "1965.5"
## [15] "1965.75" "1966" "1966.25" "1966.5" "1966.75" "1967" "1967.25"
## [22] "1967.5" "1967.75" "1968" "1968.25" "1968.5" "1968.75" "1969"
## [29] "1969.25" "1969.5" "1969.75" "1970" "1970.25" "1970.5" "1970.75"
## [36] "1971" "1971.25" "1971.5" "1971.75"Show the attributes of the dataset.
attributes(freeny)## $names
## [1] "y" "lag.quarterly.revenue" "price.index"
## [4] "income.level" "market.potential"
##
## $class
## [1] "data.frame"
##
## $row.names
## [1] "1962.25" "1962.5" "1962.75" "1963" "1963.25" "1963.5" "1963.75"
## [8] "1964" "1964.25" "1964.5" "1964.75" "1965" "1965.25" "1965.5"
## [15] "1965.75" "1966" "1966.25" "1966.5" "1966.75" "1967" "1967.25"
## [22] "1967.5" "1967.75" "1968" "1968.25" "1968.5" "1968.75" "1969"
## [29] "1969.25" "1969.5" "1969.75" "1970" "1970.25" "1970.5" "1970.75"
## [36] "1971" "1971.25" "1971.5" "1971.75"Variables in Data
Based on the findings above and general knowledge in economics, a summary of the variables can be drawn up below.
| Variable | Description | Units |
|---|---|---|
| row.names | Quarter | Year & Quarter |
| y | Quarterly revenue | $, possibly in million |
| lag.quarterly.revenue | Quarterly revenue in last quarter | $, possibly in million |
| price.index | Price index (possibly consumer price index) | Possibly % |
| income.level | Income level | Unknown |
| market.potential | Market potential | $, possibly in million |
Data Types
Check the type of the dataset. It’s found to be a list.
typeof(freeny)## [1] "list"Check the class of the dataset. It’s found to be a data frame.
class(freeny)## [1] "data.frame"3. Automated Codebook
Codebook Function
The codebook function of the memisc package is used to automatically generate a codebook of a dataset. It shows the statistics of the dataset including minimum value, maximum value, mean, standard deviation, skewness and kurtosis of the data.
codebook(freeny)## ================================================================================
##
## y
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 8.79137000
## Max: 9.79424000
## Mean: 9.30630436
## Std.Dev.: 0.31154410
## Skewness: -0.07950976
## Kurtosis: -1.25889614
##
## ================================================================================
##
## lag.quarterly.revenue
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 8.791
## Max: 9.775
## Mean: 9.281
## Std.Dev.: 0.311
## Skewness: -0.044
## Kurtosis: -1.281
##
## ================================================================================
##
## price.index
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 4.278
## Max: 4.710
## Mean: 4.496
## Std.Dev.: 0.132
## Skewness: -0.180
## Kurtosis: -1.226
##
## ================================================================================
##
## income.level
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 5.821
## Max: 6.200
## Mean: 6.039
## Std.Dev.: 0.119
## Skewness: -0.474
## Kurtosis: -1.072
##
## ================================================================================
##
## market.potential
##
## --------------------------------------------------------------------------------
##
## Storage mode: double
##
## Min: 12.970
## Max: 13.166
## Mean: 13.067
## Std.Dev.: 0.064
## Skewness: 0.024
## Kurtosis: -1.375Summary Function
Show the summary statistics of all variables in the dataset.
summary(freeny)## y lag.quarterly.revenue price.index income.level
## Min. :8.791 Min. :8.791 Min. :4.278 Min. :5.821
## 1st Qu.:9.045 1st Qu.:9.020 1st Qu.:4.392 1st Qu.:5.948
## Median :9.314 Median :9.284 Median :4.510 Median :6.061
## Mean :9.306 Mean :9.281 Mean :4.496 Mean :6.039
## 3rd Qu.:9.591 3rd Qu.:9.561 3rd Qu.:4.605 3rd Qu.:6.139
## Max. :9.794 Max. :9.775 Max. :4.710 Max. :6.200
## market.potential
## Min. :12.97
## 1st Qu.:13.01
## Median :13.07
## Mean :13.07
## 3rd Qu.:13.12
## Max. :13.174. Custom-Made Codebook
Instead of automatic generation of codebook, here the same information is shown by using individual R functions manually.
Data Types
Show the type of each variable.
sapply(freeny, typeof)## y lag.quarterly.revenue price.index
## "double" "double" "double"
## income.level market.potential
## "double" "double"Show the class of each variable.
sapply(freeny, class)## y lag.quarterly.revenue price.index
## "ts" "numeric" "numeric"
## income.level market.potential
## "numeric" "numeric"Descriptive Statistics
Show the mean of each column.
sapply(freeny, mean)## y lag.quarterly.revenue price.index
## 9.306304 9.280718 4.496182
## income.level market.potential
## 6.038596 13.066831Show the minimum value of each variable.
sapply(freeny, min)## y lag.quarterly.revenue price.index
## 8.79137 8.79137 4.27789
## income.level market.potential
## 5.82110 12.96990Show the maximum value of each variable.
sapply(freeny, max)## y lag.quarterly.revenue price.index
## 9.79424 9.77536 4.70997
## income.level market.potential
## 6.20030 13.16640Show the range of each variable.
sapply(freeny, range)## y lag.quarterly.revenue price.index income.level market.potential
## [1,] 8.79137 8.79137 4.27789 5.8211 12.9699
## [2,] 9.79424 9.77536 4.70997 6.2003 13.1664Show the first quartile of each variable.
sapply(freeny, quantile, 0.25)## y.25% lag.quarterly.revenue.25% price.index.25%
## 9.044600 9.019585 4.391615
## income.level.25% market.potential.25%
## 5.947985 13.006600Show the median of each variable.
sapply(freeny, median)## y lag.quarterly.revenue price.index
## 9.31378 9.28436 4.51018
## income.level market.potential
## 6.06093 13.06930Show the third quartile of each variable.
sapply(freeny, quantile, 0.75)## y.75% lag.quarterly.revenue.75% price.index.75%
## 9.590855 9.560515 4.604965
## income.level.75% market.potential.75%
## 6.139120 13.124400Show the interquartile range of each variable.
sapply(freeny, IQR)## y lag.quarterly.revenue price.index
## 0.546255 0.540930 0.213350
## income.level market.potential
## 0.191135 0.117800Show the variance of each variable.
sapply(freeny, var)## y lag.quarterly.revenue price.index
## 0.099613933 0.099519976 0.017784056
## income.level market.potential
## 0.014506558 0.004160757Show the standard deviation of each variable.
sapply(freeny, sd)## y lag.quarterly.revenue price.index
## 0.31561675 0.31546787 0.13335687
## income.level market.potential
## 0.12044317 0.06450393The moments package is included to show skewness and kurtosis.
# install.packages('moments') # Install the package if necessary
library(moments)Show the skewness of each variable.
sapply(freeny, skewness)## y lag.quarterly.revenue price.index
## -0.07950976 -0.04381719 -0.18010672
## income.level market.potential
## -0.47365195 0.02446065Show the kurtosis of each variable.
sapply(freeny, kurtosis)## y lag.quarterly.revenue price.index
## 1.741104 1.719270 1.774144
## income.level market.potential
## 1.928212 1.624997Visualizing Data
Histograms are generated for all variables for data visualization.
sapply(1:length(freeny), function (x) {
var_name <- names(freeny[x])
hist(freeny[[x]], main = paste0('Histogram - ', var_name), xlab = var_name, ylab = 'Frequency')
})
## [,1] [,2] [,3] [,4] [,5]
## breaks Numeric,7 Numeric,7 Numeric,11 Numeric,10 Numeric,6
## counts Integer,6 Integer,6 Integer,10 Integer,9 Integer,5
## density Numeric,6 Numeric,6 Numeric,10 Numeric,9 Numeric,5
## mids Numeric,6 Numeric,6 Numeric,10 Numeric,9 Numeric,5
## xname "freeny[[x]]" "freeny[[x]]" "freeny[[x]]" "freeny[[x]]" "freeny[[x]]"
## equidist TRUE TRUE TRUE TRUE TRUELine plots are generated for all variables for data visualization.
sapply(1:length(freeny), function (x) {
var_name <- names(freeny[x])
plot(row.names(freeny[x]), freeny[[x]], type = 'l', main = paste0('Line Plot - ', var_name), xlab = 'Quarter', ylab = var_name)
})
## [[1]]
## NULL
##
## [[2]]
## NULL
##
## [[3]]
## NULL
##
## [[4]]
## NULL
##
## [[5]]
## NULL