This site uses different types of cookies, including analytics and functional cookies (its own and from other sites). To change your cookie settings or find out more, click here. If you continue browsing our website, you accept these cookies.
You can add "No Value" to the field to replace the Nulls, but to do so you would convert the field from a number to a string. This would make the predictive model handle this data very differently and potentially disrupt how you plan to use the data. It would see the fields as categorical, no longer as continuous.
You would lose the ability to treat the field as a numeric variable; it could be treated as a categorical variable. Depending on the variable, you could bin the other values into categories. If you can set the missing flag as the reference level in your model, then the the coefficients of the other categories would be the 'effect' between each category level and 'missing'. This would be a good first look at the data.