QBoard » Statistical modeling » Stats - Tech » Metrics for Evaluating Performance of Logistic Regression

Metrics for Evaluating Performance of Logistic Regression

  • I built a Logistic Regression model that classifies whether a customer churns or not and I would like to evaluate the performance of the model. I would like to understand the best evaluation metrics for this problem.

    What do the metrics Sensitivity, Specificity, False Positives Rate, Precision, Recall, and Accuracy tell us about this model?

    This post was edited by Harish Gupta at May 5, 2019 12:17 PM IST
      May 5, 2019 12:16 PM IST
    0
  • Since Logistic regression is not same as Linear regression , predicting just accuracy will mislead. ** Confusion Matrix** is one way to evaluate the performance of your model. Checking the values of True Positives, False Negatives ( Type II Error) are really important.

    ** ROC Curve** Receiver Operating Characteristic(ROC) summarizes the model’s performance by evaluating the trade offs between true positive rate (sensitivity) and false positive rate(1- specificity)

      May 5, 2019 12:55 PM IST
    1
  • Raji Reddy A said:

    Since Logistic regression is not same as Linear regression , predicting just accuracy will mislead. ** Confusion Matrix** is one way to evaluate the performance of your model. Checking the values of True Positives, False Negatives ( Type II Error) are really important.

    ** ROC Curve** Receiver Operating Characteristic(ROC) summarizes the model’s performance by evaluating the trade offs between true positive rate (sensitivity) and false positive rate(1- specificity)

      May 5, 2019 3:52 PM IST
    0