Lastly, we can analyze how well our model performs on the test dataset.īy default, any individual in the test dataset with a probability of default greater than 0.5 will be predicted to default. Predicted <- predict(model, test, type=" response") Step 5: Model Diagnostics
LOGISTIC REGRESSION R STUDIO CODE
We can use the following code to calculate the probability of default for every individual in our test dataset: #calculate probability of default for each individual in test dataset Conversely, an individual with the same balance and income but with a student status of “No” has a probability of defaulting of 0.0439. The probability of an individual with a balance of $1,400, an income of $2,000, and a student status of “Yes” has a probability of defaulting of. Once we’ve fit the logistic regression model, we can then use it to make predictions about whether or not an individual will default based on their student status, balance, and income: #define two individuals Step 4: Use the Model to Make Predictions Since none of the predictor variables in our models have a VIF over 5, we can assume that multicollinearity is not an issue in our model. We can also calculate the VIF values of each variable in the model to see if multicollinearity is a problem: #calculate VIF values for each predictor variable in our modelĪs a rule of thumb, VIF values above 5 indicate severe multicollinearity. Balance is by far the most important predictor variable, followed by student status and then income. These results match up nicely with the p-values from the model. We can also compute the importance of each predictor variable in the model by using the varImp function from the caret package: caret::varImp(model) We can compute McFadden’s R 2 for our model using the pR2 function from the pscl package: pscl::pR2(model)Ī value of 0.4728807 is quite high for McFadden’s R 2, which indicates that our model fits the data very well and has high predictive power. In practice, values over 0.40 indicate that a model fits the data very well. Values close to 0 indicate that the model has no predictive power. Instead, we can compute a metric known as McFadden’s R 2, which ranges from 0 to just under 1. However, there is no such R 2 value for logistic regression. This number ranges from 0 to 1, with higher values indicating better model fit. In typical linear regression, we use R 2 as a way to assess how well a model fits the data. We can see that balance and student status seem to be important predictors since they have low p-values while income is not nearly as important. The p-values in the output also give us an idea of how effective each predictor variable is at predicting the probability of default: For example, a one unit increase in balance is associated with an average increase of 0.005988 in the log odds of defaulting. The coefficients in the output indicate the average change in log odds of defaulting. Residual deviance: 1065.4 on 6960 degrees of freedom
Null deviance: 2021.1 on 6963 degrees of freedom (Dispersion parameter for binomial family taken to be 1) We can use the following code to load and view a summary of the dataset: #load dataset Step 1: Load the Dataįor this example, we’ll use the Default dataset from the ISLR package.
LOGISTIC REGRESSION R STUDIO HOW TO
This tutorial provides a step-by-step example of how to perform logistic regression in R. We then use some probability threshold to classify the observation as either 1 or 0.įor example, we might say that observations with a probability greater than or equal to 0.5 will be classified as “1” and all other observations will be classified as “0.” Thus, when we fit a logistic regression model we can use the following equation to calculate the probability that a given observation takes on a value of 1: The formula on the right side of the equation predicts the log odds of the response variable taking on a value of 1. β j: The coefficient estimate for the j th predictor variable.Logistic regression uses a method known as maximum likelihood estimation to find an equation of the following form: Logistic regression is a method we can use to fit a regression model when the response variable is binary.
0 kommentar(er)
