0
$\begingroup$
  • I am trying to build a model where my target is continuous with a right-skewed distribution. I am interested to know what are possible models I can try apart from linear regression.
  • I have around 200 Predictors(mostly numeric and few categorical) and 45,000 records.
  • Target Ranges from 10-100 (80% sample less has target value than 50). My goal is to predict either 'exact value' or in the correct band(10-15, 15-20,20-25..95-100).

Thank you

$\endgroup$
2
  • 2
    $\begingroup$ All of the supervised learning regression methods are viable: regularized regression, random forest, support vector regression, neural networks. However, what is your goal in fitting the regression? That may help constrain the problem. $\endgroup$
    – Dave
    Apr 25 at 16:53
  • $\begingroup$ Apart from indicating the response is continuous and you have a dataset of moderate size, you haven't supplied any information that would distinguish any particular method as more or less appropriate. $\endgroup$
    – whuber
    Apr 25 at 18:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.