Suppose we have cost data on some product line, let's say clothing. A simple model of the true cost of the clothing could look something like this:

C$_{i,j,k}$ = c$_i \cdot \alpha_j \cdot \beta_k $,

for some specific article of clothing, $i$, made in some month, $j$, in some region $k$. Then c$_i$ is the true cost, $\alpha_j$ is a factor for the month and $\beta_k$ is a factor for the region.

The question is how could we estimate the parameters of this model? I am having lots of trouble setting something up. I am only somewhat familiar with Maximum Likelihood Estimation, but I have no idea how the Likelihood would look like in the first place. How does this multi-dimensional aspect of the model get taken care of?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.