## How to calculate MAPE & MSE according to two columns of data

Regular Contributor
Posts: 154

# How to calculate MAPE & MSE according to two columns of data

Hi,

I have a dataset with name 'Data_1' with four columns: 'Factory', 'Item', 'Date', 'Actual_Demand' & 'Prediction_Demand'.

I want to calculate MAPE and MSE that 'Prediction' via 'Actual' group by 'Factory' & 'Item' across 'Date'.

And my code is:

Proc Sql;
Create Table Work.Result_1 as
select distinct
Factory,
Item,
avg(abs((Actual_Demand - Prediction_Demand)/Actual_Demand)) as MAPE,
avg((Prediction_Demand - Actual_Demand)**2) as MSE
from Work.Data_1
group by 1, 2;
Quit;

So that I could obtain a new table with two new columns. Is the formula I have in the code correct to calculate 'MAPE' & 'MSE'?

Thank you very much!

Regular Contributor
Posts: 154

## MSE & MAPE Function correct?

Hi,

I have a dataset with name 'Data_1' with four columns: 'Factory', 'Item', 'Date', 'Actual_Demand' & 'Prediction_Demand'.

I want to calculate MAPE and MSE that 'Prediction' via 'Actual' group by 'Factory' & 'Item' across 'Date'.

And my code is:

Proc Sql;
Create Table Work.Result_1 as
select distinct
Factory,
Item,
avg(abs((Actual_Demand - Prediction_Demand)/Actual_Demand)) as MAPE,
avg((Prediction_Demand - Actual_Demand)**2) as MSE
from Work.Data_1
group by 1, 2;
Quit;

So that I could obtain a new table with two new columns. Is the formula I have in the code correct to calculate 'MAPE' & 'MSE'?

Thank you very much!

Discussion stats