I have been working on creating a strategy and forecasting eletricity prices. My problem is that I think I confused my self in the process. The residuals of the series looks normally distributed and centered around 0 which fits the model assumptions. But why do the residuals need to be fullfill these properties?
I read somewhere that the data is random when these assumptions aren't broken. But I don't understand why it would be a problem that the data is not random and is systematic. In my head I thought the process is about wanting to see trends and systems in the data that you can use for your benefit.
Can you explain why I want random data? Maybe help clear these confusions and tell me how I would be able to use random data to forecast?
- 267 views
- Number of different drinks that can be made using 6 ingredients
- Sample size calculation for a cross sectional healthcare study
- Bayesian Statistics - Zero Inflated Binomial Model - Calculate Posterior Conditional Distribution
- Ridge Regression for Panel Data Analysis
- 2 conceptual college statistics questions - no equations
- CLT and probability
- Statistics for argumentative data and politics
- Normal distribution & Probability