Exam DP-100 topic 1 question 24 discussion

Actual exam question from Microsoft's DP-100

Question #: 24
Topic #: 1

DRAG DROP -
You are in the process of constructing a regression model.
You would like to make it a Poisson regression model. To achieve your goal, the feature values need to meet certain conditions.
Which of the following are relevant conditions with regards to the label data? Answer by dragging the correct options from the list to the answer area.
Select and Place:

Show Suggested Answer

Suggested Answer:

Poisson regression is intended for use in regression models that are used to predict numeric values, typically counts. Therefore, you should use this module to create your regression model only if the values you are trying to predict fit the following conditions:
✑ The response variable has a Poisson distribution.
✑ Counts cannot be negative. The method will fail outright if you attempt to use it with negative labels.
✑ A Poisson distribution is a discrete distribution; therefore, it is not meaningful to use this method with non-whole numbers.
Reference:
https://docs.microsoft.com/en-us/azure/machine-learning/studio-module-reference/poisson-regression

by pancman at April 13, 2022, 6:57 p.m.

Comments

Submit Cancel

pancman

Highly Voted 2 years, 9 months ago

Here are the requirements of the Poisson regression: The response variable has a Poisson distribution. Counts cannot be negative. The method will fail outright if you attempt to use it with negative labels. A Poisson distribution is a discrete distribution; therefore, it is not meaningful to use this method with non-whole numbers.

upvoted 13 times

...

pancman

Highly Voted 2 years, 9 months ago

Answer is correct

upvoted 7 times

...

evangelist

Most Recent 11 months, 1 week ago

For constructing a Poisson regression model, the label (dependent variable) must meet specific conditions due to the nature of Poisson distribution. Poisson regression is used for modeling count data and the events occurrence in a fixed interval of time or space. The correct conditions for the label data in a Poisson regression model are: Explanation: It must be whole numbers: Poisson regression models the count of occurrences of an event, thus the dependent variable must indeed be whole numbers (integers) since you cannot have a fraction of an occurrence. It must be a positive value: More precisely, the values must be non-negative integers (including zero). Poisson regression is suitable for counts which cannot be negative.

upvoted 1 times

...

james2033

1 year, 3 months ago

* It must be whole numbers. * It must be a positive value.

upvoted 3 times

...

phydev

1 year, 6 months ago

Yep, Poisson regression do be like that.

upvoted 3 times

...