Lecture_Forecast%20Error

Lecture_Forecast%20Error - Forecast Error Accuracy and Bias...

Info iconThis preview shows pages 1–9. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Forecast Error Accuracy and Bias Concepts No Bias, Poor Accuracy Bias, Poor Accuracy Bias, Good Accuracy No Bias, Good Accuracy Determining Forecast Accuracy and Bias • Post-Hoc (after the fact) analysis • Comparing a forecast with a known observation – Census – Estimate • Evaluating previous forecasts • Forecast experiments Measures of Forecast Error 10-Year Forecast Errors, Selected Counties in Florida 1 2000 County Census (A) Forecast (F) E abs(E) PE abs(PE) (PE) 2 BAY 148,217 147,684-533 533-0.4 0.4 0.1 BREVARD 476,230 504,263 28,033 28033 5.9 5.9 34.7 BROWARD 1,623,018 1,469,802-153,216 153216-9.4 9.4 89.1 DADE 2,253,362 2,173,098-80,264 80264-3.6 3.6 12.7 DE SOTO 32,209 28,290-3,919 3919-12.2 12.2 148.0 FLAGLER 49,832 46,423-3,409 3409-6.8 6.8 46.8 HILLSBOROUGH 998,948 970,405-28,543 28543-2.9 2.9 8.2 JEFFERSON 12,902 13,169 267 267 2.1 2.1 4.3 MANATEE 264,002 260,334-3,668 3668-1.4 1.4 1.9 MONROE 79,589 90,470 10,881 10881 13.7 13.7 186.9 NASSAU 57,663 53,016-4,647 4647-8.1 8.1 64.9 SARASOTA 325,957 339,398 13,441 13441 4.1 4.1 17.0 SUMTER 53,345 37,491-15,854 15854-29.7 29.7 883.3 SUWANNEE 34,844 30,921-3,923 3923-11.3 11.3 126.8 WAKULLA 22,863 17,052-5,811 5811-25.4 25.4 646.0-16,744 23,761-5.7 9.1 12.3 ME MAE MALPE MAPE RMSPE-3,919 5,811-3.6 6.8 MEDE MEDAE MEDALPE MEDAPE 1 Forecast based on cohort component model and produced in 1991 % Positive 26.7% Average Median Forecast Error Distributions Positive-Skew Average and Median Absolute Percent Errors in 1990, Selected Counties and Census Tracts Counties MAPE MEDAPE RATIO Arkansas 4.1 3.6 1.14 Florida 9.8 7.6 1.29 Iowa 8.0 7.8 1.03 North Dakota 19.2 17.7 1.08 Ohio 3.2 2.4 1.33 Washington 4.3 4.0 1.08 Wyoming 15.1 10.4 1.45 Census Tracts MAPE MEDAPE RATIO Detroit 28.5 11.2 2.54 Dallas-Ft. Worth 18.6 11.5 1.62 San Diego 20.9 14.5 1.44 Computing Naïve Forecast Error 10-Year Naïve Forecast Errors, Selected Counties in Florida 1 2000 Jurisdiction Census (A) Naive (F) PE abs(PE) BAY 148,217 140,216-5.4% 5.4% BREVARD 476,230 455,230-4.4% 4.4% BROWARD 1,623,018 1,463,281-9.8% 9.8% DADE 2,253,362 2,334,730 3.6% 3.6% DE SOTO 32,209 28,764-10.7% 10.7% FLAGLER 49,832 34,593-30.6% 30.6% HILLSBOROUGH 998,948 964,323-3.5% 3.5% JEFFERSON 12,902 13,615 5.5% 5.5% MANATEE 264,002...
View Full Document

This note was uploaded on 10/20/2010 for the course ECON 125 taught by Professor Tayman during the Spring '08 term at UCSD.

Page1 / 32

Lecture_Forecast%20Error - Forecast Error Accuracy and Bias...

This preview shows document pages 1 - 9. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online