Get certified for free when you join Fabric Data Days 2026 and dive into Fabric, Power BI, SQL, AI, and other essential data skills.
Join nowJuly 7 - July 17 | Round 2 of the Power BI Dataviz World Championships. Don't miss your chance! Learn more
Hello,
I am trying to forecast sales using an R script. The R script works fine in R Studio but returns an empty table when used in Power BI.
Here is my data:
| Week | Unit |
| 1/3/2018 | 355263 |
| 1/10/2018 | 398547 |
| 1/17/2018 | 332485 |
| ... | ... |
"Week" is formatted as a date and "Unit" is formatted as a number.
I use the following code:
library(dplyr)
library(forecast)
library(lubridate)
salesTS <- ts(dataset$`Unit`, frequency = 52, start = 2018)
salesLog <- log(salesTS)
salesLogHW <- HoltWinters(salesLog)
nextYearSales <- forecast(salesLogHW, h=52)
"This table is empty" is displayed after I try to transform the dataset to time series.
Does anyone know how I can make my code work and obtain a forecast for 52 weeks ahead?
Thanks,
Alona
where are the week dates in your R script? (the "week" column)
Join us in Barcelona for FabCon and SQLCon, the Fabric, Power BI, SQL, and AI community event. Save €200 with code FABCMTY200.
Join Data Days 2026: 60 days of free live/on-demand sessions, challenges, study groups, and certification opportunities.
| User | Count |
|---|---|
| 27 | |
| 24 | |
| 18 | |
| 15 | |
| 15 |
| User | Count |
|---|---|
| 63 | |
| 42 | |
| 41 | |
| 38 | |
| 37 |