# Should I transform a multiple regression with outliers into ordinal regression?

I have small dataset of about 60 samples that performs poorly in regression. So I wonder how can I transform this task into predicting intervals instead of values. Is it possible to make it perform better for intervals instead of values?

The dataset is imbalanced. There are just few high values. The algorithm constantly underestimates these values. So I think if I make prediction in intervals, it'll perform better for values, say, more than 50.

But, overall, the problem is predicting for values near the limits of intervals. Is it better to mark this values as belonging to both intervals?

Your final proposal (put an observations "on the edge" of intervals into both intervals) is unconventional, but probably worth trying. Note that ordinal logistic regression need more parameters than linear regression, so be careful adding too many categories because your data is small.

You can consider transforming the prediction target to reduce the distance between the extreme high values and the rest of the data.

In general, however, you will need to understand why those extreme high values are high. Are they incorrect entries (i.e. outliers)? Or is there some feature in the data that can be added to the model in order to better predict when a data point is higher than expected?

For example:

Let's say you are trying to predict student performance on a linear algebra midterm exam. Your data set consists of last year's exam scores. Maybe your model includes these features:

• Student class year (freshman, sophomore, junior, senior)
• Did student take honors calculus? (binary)