Reaching on this one, but in the context of supply chain MAPE, why do we use shipments as the denominator instead of forecast?

I thought this was your field of expertise.

Why are you asking for help on this topic?

This is a statistical theory question so you won't find any hall pass violations here.


Nothing to see, move a long, you're lost.


I can just hear chunk thinking right now . . . .what witty post can I make?

Well, why do you think you use shipments?

I don't use MAPE when talking about data with real-life scientists. P-values, confidence intervals and study design are what people I work with need to know in order to do their jobs better.

So, are you saying you need help with how to do your job? There is software that can help you if you are worried about your skill set. A monkey can set up a supply chain with some of the tools available these days. It's not my arena, but I have a friend that consults some CA manufacturers on getting raw materials. I can put you in contact with him if this is a legitimate question.

My two cents is that shipments are raw data while forecasts are speculatory, hence using actual historical numbers to base your supply chain on instead of the unknown future, but again, I thought this was your wheelhouse.

Not statisical theory, just basic math.

MAPE measures the relationship of the forecast to actual........actual (beginning) is always the base amount. Kind of like calculating a rate of return or change.



If you start at 5 (actual) and increase to 10 (forecast) your rate is 100%

(10-5)/5 = 1

If you use the ending amount (or forecast) as the base it gives an incorrect rate (10-5)/10 = .5

Sheik???

He's posting questions about Palestine on a statistics board. Give him some time.

A simple "thank you for explaining a detail of my job that I never understood; now I am slightly ahead of all the other button pushers" would be nice....... :<_<:

Dude, he's busy. He's looting Arafat's old compound.

Taking a breather in an armchair while others fight came all too naturally for him......

On my dilznik.

Thanks for the interest.

Nice try though Space Ape, answer the why, in the context of a supply chain.

The 8th grade math calculation was . . .neat.

Why are actuals the divisor when calculating error?

:spaz:

Internet Math Toughguy Alert! On my dilznik. Good.

In context:
shipments = beginning
forecast = ending

You want your forecast to be shown relative to actuals.....therefore actuals are the denominator.

It's not just specific to supply chain, and the logic goes even beyond mape calcs. Your base amount is always the historical/actual/beginning value; that simple. MAPE isnt unique to supply chain management so neither is the calculation of it. I'm not going to waste time trying to explain rates of change to you a hundred different ways.

Think about what you're trying to accomplish with MAPE; its merely a relationship measure ie rate. There's no other way to explain it because it really is that simple.

To avoid bias by the forecaster. Higher forecast =lower MAPE

Thats the general use of mape, not the logic for the actual structure of the formula. BTW, that second part isnt always going to be true.

MAPE is for pussies.

MAPE ape, MAPE ape

Sheik, are you free-lancing as a MAPE consultant these days?