Introduction

The linear model still remains a reference point towards advanced modeling of some datasets as foundation for Machine Learning, Data Science and Artificial Intelligence in spite of some of her weaknesses. The major task in modeling is to compare various models before a selection is made for one or for advanced modeling. Often, some trial and error methods are used to decide which model to select. This is where this function is unique. It helps to estimate 14 different linear models and provide their coefficients in a formatted Table for quick comparison so that time and energy are saved. The interesting thing about this function is the simplicity, and it is a one line code. The different transformations are:

• Linear model

• Linear model with interactions

• Semilog model

• Growth model

• Double Log model

• Mixed-power model

• Translog model

• Quadratic model

• Cubic model

• Inverse of y model

• Inverse of x model

• Inverse of y & x model

• Square root model

• Cubic root model

In this blog, I share with you a function Linearsystems from Dyn4cast package which you can use in a very easy way to perform linear regression and or its various transformations. It is a one line code and easy to use. The usage is as follows:

Linearsystems(y, x, mod, limit, Test = NA)

y is the vector of the dependent variable.

x is the vector of the independent variables preferable in data.frame.

mod is the group of linear models to be estimated. It takes value from 0 to 6. 0 = EDA (correlation, summary tables, Visuals means); 1 = Linear systems, 2 = power models, 3 = polynomial models, 4 = root models, 5 = inverse models, 6 = all the 14 models.

limit is the number of variables to be included in the coefficients plots.

Test is the test data to be used to predict y. If not supplied, the fitted y is used hence may be identical with the fitted value.

With this one line of codes, in addition to the individual estimated models, the following are what you get:

Visual means of the numeric variable

Correlation plot

Significant plots of all the models estimated

Model Table

Machine Learning Metrics which is also a list of 47 performance and diagnostic statistic

Table of Marginal effects

Fitted plots long format

Fitted plots wide format

Prediction plots long format

Prediction plots wide format

Naive effects plots long format

Naive effects plots wide format

Summary of numeric variables

Summary of character variables

Let us dive into an awesome experience in machine learning!

Load library

library(Dyn4cast)

Estimate without test data

y <- linearsystems$MKTcost
x <- select(linearsystems, -MKTcost)
Model1 <- Linearsystems(y, x, 6, 15)

Model1$`Correlation plot`$plot()

Model1$`Model Table`

Model1$`Significant plot of Double Log`

Model1$`Fitted plots wide format`

Model1$`Tables of marginal effects`[[1]]

Model1$`Naive effects plots long format`

Estimation with test data

x <- sampling[, -1]
y <- sampling$qOutput
Data <- cbind(y, x)
sampling <- sample(1:nrow(Data), 0.8 * nrow(Data)) # 80% of data is sampled for training the model
train <- Data[sampling, ]
Test <- Data[-sampling, ] # 20% of data is reserved for testing (predicting) the model
y <- train$y
x <- train[, -1]
mod <- 4
Model2 <- Linearsystems(y, x, 4, 15, Test)

Model2$`Model Table`

Model2$`Table of Marginal effects`

Model2$`Visual means of the numeric variable`

Model2$`Fitted plots long format`

Model2$`Prediction plots long format`

Model2$`Significant plot of Square root`

Model2$`Machine Learning Metrics`