#Logistic regression in r studio how to#
#Logistic regression in r studio code#

In the following code you change the level as follow: Old level It is also possible to create lower levels for the marital status. Summarize(average_educ_year = mean(educational.num), In the table below, you create a summary statistic to see, on average, how many years of education (z-value) it takes to reach the Bachelor, Master or PhD. We change the values of education with the statement ifelse We use the verb mutate from dplyr library.Ifelse(education = 'Masters' | education = 'Prof-school', 'Master', 'PhD'))))))) Mutate(education = factor(ifelse(education = 'Preschool' | education = '10th' | education = '11th' | education = '12th' | education = '1st-4th' | education = '5th-6th' | education = '7th-8th' | education = '9th', 'dropout', ifelse(education = 'HS-grad', 'HighGrad', ifelse(education = 'Some-college' | education = 'Assoc-acdm' | education = 'Assoc-voc', 'Community', Higher levels of education will be changed to master. For instance, low level of education will be converted in dropout. Namely, you create larger groups with similar level of education. If you want to improve the amount of information you can get from this variable, you can recast it into higher level. This is substantial, and some levels have a relatively low number of observations. Step 3) Feature engineering Recast educationįrom the graph above, you can see that the variable education has 16 levels. Note: Use the next button to navigate to the next graph Note, to return x as a column, you need to include it inside the get() ggplot(factor, aes(get(x))) + geom_bar()+ theme( = element_text(angle = 90)): Create a bar char chart for each x element.function(x): The function will be processed for each x.lapply(): Use the function lapply() to pass a function in all the columns of the dataset.It is more convenient to automatize the process, especially in situation there are lots of columns. You want to plot a bar chart for each column in the data frame factor. The dataset contains 6 categorical variables The library ggplot2 requires a data frame object. ame(select_if(data_adult, is.factor)): We store the factor columns in factor in a data frame type.We can select the factor columns with the code below: # Select categorical columnįactor <- ame(select_if(data_adult_rescale, is.factor)) Store the bar chart of each column in a list.

We will divide this step into three parts:

Check the level in each categorical column.Output: # X age workclass education educational.num mutate_if(is.numeric, funs(scale)): The condition is only numeric column and the function is scale.Mutate_if(is.numeric, funs(as.numeric(scale(.)))) In the first step, you can see the distribution of the continuous variables. In this tutorial, each step will be detailed to perform an analysis on a real dataset. Your task is to predict which individual will have a revenue higher than 50K. Step 7: Assess the performance of the model.$ gender Male, Male, Male, Male, Female, Male, Male, Male. $ race Black, White, White, Black, White, White, Black. $ marital.status Never-married, Married-civ-spouse, Married-civ-sp. $ education 11th, HS-grad, Assoc-acdm, Some-college, Some-col. $ workclass Private, Private, Local-gov, Private, ?, Private. marital.status: Marital status of the individual.education: Educational level of the individual.The dataset contains 46,033 observations and ten features: The objective is to predict whether the annual income in dollar of an individual will exceed 50.000. The “adult” is a great dataset for the classification task. Let’s use the adult data set to illustrate Logistic regression.

#Logistic regression in r studio how to#

How to create Generalized Liner Model (GLM)

Step 7) Assess the performance of the model.

How to create Generalized Liner Model (GLM).

All values above this threshold are classified as 1 To convert a continuous flow into discrete value, we can set a decision bound at 0.5.