1 / 14

Impact of Rounding on Sample Estimate Accuracy in ERCOT Load Profiling

This study examines the influence of rounding on the accuracy of sample estimates and load profiling in ERCOT. Monte Carlo simulations are conducted to analyze the impact of different levels of rounding on sample precision.

mistier
Télécharger la présentation

Impact of Rounding on Sample Estimate Accuracy in ERCOT Load Profiling

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Impact of Sample Estimate Rounding on Accuracy ERCOT Load Profiling Department May 22, 2007

  2. Overview • Objective is to assess the impact of rounding on the accuracy of sample estimates and on load profile accuracy • Monte Carlo simulation of sample results • Settlement in a test environment

  3. Simulation Methodology • Monte Carlo simulations of repeated sampling to analyze the impact of various levels of rounding on the precision of the resulting sample estimates • Use SAS random number generator functions • RANUNI … random numbers from a uniform distribution to generate test population means • RANNOR … random numbers from a normal distribution to generate sample estimates based on a population mean and standard deviation

  4. Simulation Methodology (continued) • Simulation steps • Generate a randomized population mean • Simulate a sample result based on • Sampling from a population having that mean • Based on a sample design with a selected statistical accuracy (at 90% confidence level) • Round the sample estimate to the hundredths, thousandths and ten-thousandths place • Calculate the difference between the rounded estimate and the population mean • Replicate 10,000 times at each precision level

  5. Simulation Results

  6. Simulation Results

  7. Simulation Results

  8. Simulation Results

  9. Simulation Results

  10. Simulation Findings • If the precision is ±10% or worse, and the population mean is > 0.3, MAPEs for 2-digit rounding and 3-digit rounding are virtually identical • If the precision is better than ±10%, and the population mean is < 0.3, MAPEs for 3-digit rounding are somewhat better • If the population mean is ≥ 1, 3-digit rounding is likely to do more harm than good regardless of the sample precision

  11. Test Settlement • Data Aggregation ran a test settlement using profiles with three digit rounding for January 27, 2006 • Compared UFE for original 2-digit profiles with results for 3-digit profiles • Compared total aggregated residential load for the same day

  12. January 27, 2006 UFE Comparison

  13. January 27, 2006 Residential Load Comparison Red = 3 decimals Blue = 2 decimals

  14. Test Settlement Result • 3-digit profile rounding produced virtually no difference in settlement for the selected day

More Related