Reputation: 15
I would appreciate your opinion on this. I usually work my data on MS Excel or Python to get an output. For instance, I use daily prices to calculate daily, montly, quarterly retruns, then calculate standard deviation and so on. Now, I need to show those outputs or calculations derived from a price series to PowerBI. Is DAX on PowerBI enough to do any kind of calculations begining with a daily price serie, or I should stick with Python or Excel to do the math and then show it through PowerBI.
For instance a mutual fund screener with an only input such as the daily price (leaving all qualitative aside) showing, montly returns, histograms, standard deviation graph evolution, and so on with another risk metrics.
Thanks any opinion or reference will be welcomed! best practices, reading material, etc.
Upvotes: 0
Views: 46
Reputation: 145
DAX is a pretty powerful tool and can likely do what you need. Generally best practice would be to have all of the raw data as clean as possible once it gets to PowerBI if possible (depending on where it's coming from). For instance if you're importing data from a server and do transformations in Query Editor, you're going to run into issues if the dataset is very large. Query Editor is executed on your machine while SQL would be executed on the server, so it can run much slower. You can get around this by cleaning the data before it gets sent to PowerBI or by query folding. If you go to the data tab in the sidebar on PowerBI desktop, you can add measures and columns for what you want to do. For statistical calculations such as standard deviation you can look here: https://learn.microsoft.com/en-us/dax/statistical-functions-dax
Upvotes: 0