Predicting with logistic regression
The next step is to do some analytics also in R. I will start with logistic regression. First, we need to read the data in R:
library(RODBC)
con <- odbcConnect("AWDW", uid = "RUser", pwd = "Pa$$w0rd")
TM <-
sqlQuery(con,
"SELECT CustomerKey, CommuteDistance,
TotalChildren, NumberChildrenAtHome,
Gender, HouseOwnerFlag,
NumberCarsOwned, MaritalStatus,
Age, Region, Income,
Education, Occupation,
BikeBuyer, TrainTest
FROM dbo.TMTrain
UNION
SELECT CustomerKey, CommuteDistance,
TotalChildren, NumberChildrenAtHome,
Gender, HouseOwnerFlag,
NumberCarsOwned, MaritalStatus,
Age, Region, Income,
Education, Occupation,
BikeBuyer, TrainTest
FROM dbo.TMTEST")
close(con)The next step, similar to Python, is to define the factors and their levels accordingly:
# Define Education as ordinal
TM$Education = factor(TM$Education, order = TRUE,
levels = c("Partial High School",
"High School", "Partial College",
"Bachelors", "Graduate Degree"))
# Define...
