November 26, 2017 at 11:26 pm
Hi,
I'm trying to create a forecast model based on sales figures. But I've ran into a couple of issues.
I've followed a couple of videos - presented by MVP's - based on forecasting, but I'm not able to do what they're doing using my own data.
I'm using Visual Studio 2013 Ultimate addition, and is deploying the data model (or trying to deploy) to a SQL 2014 enterprise edition server. I'm using the Microsoft Time Series model for this.
We have a sales hierarchy within the organisation of - Region/Branch/Sales Team, and the forecasting must be done on each level.
My base table that contains all the data is called AgentDataForeCasts and have the following columns:
SalesMonth - Date - This is the date/month of when the previous sales were made
Region - Varchar - Region of sale
BranchName - Varchar - Branch that made the sale
TeamName - Varchar - Name of the team that done the sale
TotalSalesPersons - int - This is the total of Sales people that worked within the Team. (Active Sales people within SalesMonth, this can change from month-to-month as people resign/hired)
TotalSalesMade - int - The total sales made within the team.
ok - my first issue is:
Can this be done somehow? Are there any settings that I've missed?
The second issue is, and one that I simply can't get around is that I'm getting this error when I'm trying to deploy the model:
Error 1 Error (Data mining): The BranchName column of the Agent Data Check Fore Cast model contains a content type that is not supported by the algorithm. The BranchName is a varchar - the Content property is set to Discrete. The total number of branches is 88 - and no null values within this column.
Not sure what I'm missing here....
Thanks in advance!
Viewing 0 posts
You must be logged in to reply to this topic. Login to reply