HR Analytics Assignment: To Find

You might also like

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 3

HR Analytics Assignment

To Find –

1. Key drivers of renege


2. Set the rules (Based on the method you have decided, Logistic or Decision tree or
any other)
3. Predictive model to predict renege.

Findings –

Key drivers of renege – After the Information Value was computed following three
relevant drivers came significant.

Variable Name
Notice period
LOB
Duration to
accept the offer

Setting the rules -

RStudio was used to do the exploratory data analysis. R script was written to develop
the logistic regression model and to compute the Information Value of all the variables.

Information Value Script –

##Information Value
install.packages("devtools")
library(devtools)
install_github("tomasgreif/riv")
library(woe)
iv <- iv.mult(Data_Frame_Scalene[,!names(Data_Frame_Scalene) %in% c("Candidate
Ref")],"Status",TRUE, verbose = FALSE)
iv.plot.summary(iv)
iv
Logistic Regression Script –

##Logistic Regression

mylogit <- glm( formula = Status ~ LOB + DTAO + NP , data = LR_DF[-1], family =
“binomial")
mylogit

Predictive model to predict renege – Logistics Regression

Output –
Coefficients:
(Intercept) LOBBFSI LOBCSMP LOBEAS LOBERS
2.121063 -0.099707 0.161759 -0.233103 -0.032139
LOBETS LOBHealthcare LOBINFRA LOBMMS DTAO
0.337849 0.297030 0.609902 12.437495 0.002236
NP
-0.020579

Degrees of Freedom: 9009 Total (i.e. Null); 8999 Residual


Null Deviance: 8683
Residual Deviance: 8251 AIC: 8273

The predicted Renege Status is given by :

Status =
Intercept * 2.121063 +
LOBBFSI * -0.099707 +
LOBCSMP * 0.161759 +
LOBEAS * -0.233103 +
LOBERS * -0.032139 +
LOBETS * 0.337849 +
LOBHealthcare * .29703 +
LOBINFRA * 0.609902 +
LOBMMS * 12.437495 +
DTAO * 0.002236 +
NP * -0.020579

Inference –

For the negative coefficient values, the value of renege decreases as the independent
variable increases. For the positive coefficient variables, the amount of renege and the
independent variable move in the same direction.

You might also like