Forecast Verification

Forecast Verification Presenter: Neil Plummer National Climate Centre Lead Author: Scott Power Bureau of Meteorology Research Centre Acknowledgements A. Watkins, D. Jones, P. Reid, NCC Introduction  Verification - what it is and why it is important?  Terminology  Potential problems  Comparing various measures  Assisting users of climate information What is verification?  “check truth or correctness of”  “process of determining the quality of forecasts”  “objective analysis of degree to which a series of forecasts compares and contrasts with the equivalent observations of a given period” Why bother with verification?  Scientific admin support o is a new system better? o assist with consensus forecasts  Application of forecasts o “how good are your forecasts?” o “should I use them?” o can be used to help estimate value Terminology can be confusing  Verification is made a little tricky by the fact that everyday words are used to describe quantities with a precise statistical meaning. Common words include: o accuracy o skill o reliability o bias o value o hit rates, percent consistent, false alarm rate, ...  all have special meanings in statistics Accuracy  Average correspondence between forecasts and observations  Measures o mean absolute error, root mean square error Bias  Correspondence between average forecast with average observation o e.g. average forecast - average value of observation Skill  Accuracy of forecasts relative to accuracy of forecasts using a reference method (e.g. guessing, persistence, climatology, damped persistence, …)  Measures o numerous! Reliability  Degree of correspondence between the average observation, given a particular forecast, and that forecast taken over all forecasts  e.g. suppose forecasts of : “10% or 30% or , …, or 70% or … chance of rain tomorrow” are routinely issued for many years  if we go back through all of the forecasts issued a forecast of looking for occasions when forecast probability of 70% was issued, then we would expect to find rainfall on 70% of occasions if the forecast system is “reliable”  this is often not the case Reliability Graph Reliability Graph 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Reliability Seasonal Forecast Reference Probability Value  Impact that prudent use of a given forecast scheme has on the user’s profits, COMPARED WITH profits made using a reference strategy  Measures o $, lives saved, disease spread reduced, … Contingency Table FORECAST YES NO Miss OBSERVED YES Hit NO False Alarm Correct Rejection HIT RATE = Hits/(Hits + Misses) FALSE ALARM RATE = False Alarms/(False Alarms + Correct Rejections) PERCENT CONSISTENT = 100*(Hits+Correct Rejections)/Total Accuracy measures  Hit rates o Proportion of observed events correctly forecast  False alarm rates o Proportion of observed non-events forecasted as events  Percent Correct o 100x (proportion of all forecasts that are correct) 1. Forecast performance 2x2 contingency table Forecast event Event 28 72 100 Nonevent 23 2680 2703 Total 51 2752 2803 Obs Nonevent Total Is this a good scheme?  1. Original Scheme: Percent correct = 100(28 + 2680)/2803 = 96.6% so it is a very accurate scheme! or is it? 2. Performance of 2nd (reference) forecast method: never predict a tornado = a “lazy” forecast scheme! Forecast event Event Obs 0 0 0 Nonevent 51 2752 2803 Total 51 2752 2803 Nonevent Total Performance measures Percent Correct:  1. Original Scheme: Percent correct = 100(28 + 2680)/2803 = 96.6%  2. Reference Lazy Scheme: Percent correct = 100(0 + 2752)/2803 = 98.2% !! Performance measures Hit rates: 1 ) 28/51 … so over half the tornadoes predicted 2 ) reference scheme: 0/51 … no tornadoes predicted Value  Suppose an unexpected (unpredicted) tornado causes $500 million damage and that an expected (predicted) tornado results in only $100 million damage  So forecast scheme (1) saves 28 x 400 million compared to forecast scheme (2)  a huge saving - highly valuable!! Categorical versus probabilistic  Categorical o “The temperature will be 26ºC tomorrow”  Probabilistic o “There is a 30% chance of rain tomorrow” o “There is a 90% chance that wet season rainfall will be above median” Artificial Skill  danger of too many inputs  danger of trying too many inputs  independent data  cross-validation  importance of supporting evidence o simple plausible hypothesis o climate models o process studies How do users verify predictions?  No single answer, however: o some switch from probabilistic to categorical o media prefer categorical forecasts o assessments made on a single season o extrapolation How can we assist users in verification  Increase access to verification information  Simplify information  Build partnerships o media o users & user groups o other government departments  Education (booklets, web, …) Summary  Verification is crucial but care is needed!  Familiarise with terminology used o skill, accuracy, value, …     No single measure tells the whole story Importance of using independent data in verification Keep it simple Communicating verification results is challenging o Users sometimes do their own verification - sobering o Most people like to think categorically - challenging o Dialogue with end-users is very important

Related docs
Verification of Probabilistic Forecast
Views: 3  |  Downloads: 1
Forecast Verification Workshop
Views: 1  |  Downloads: 0
Fuzzy Forecast Verification
Views: 0  |  Downloads: 0
Forecast Verification Report
Views: 4  |  Downloads: 1
Probabilistic Forecast Verification
Views: 0  |  Downloads: 0
Fuzzy Forecast Verification
Views: 1  |  Downloads: 0
Categorical Forecast Verification
Views: 0  |  Downloads: 0
A General Framework for Forecast Verification
Views: 0  |  Downloads: 0
NHC Forecast Verification
Views: 0  |  Downloads: 0
Other docs by Kimberly Brozi...
Cover Letter to US Patent and Trademark Office
Views: 279  |  Downloads: 5
I Have A Dream Speech
Views: 408  |  Downloads: 8
Board Resolution For Leasing Corporate Offices
Views: 255  |  Downloads: 4
2007 Inst W-2 and W-3 (PDF) Instructions
Views: 248  |  Downloads: 1
Customer Service Action Form
Views: 552  |  Downloads: 16
Employee Application For Prospective Employees
Views: 259  |  Downloads: 8
Accenture Partnership Agreement
Views: 285  |  Downloads: 13
Caldera Systems Inc Ammendments and By laws
Views: 164  |  Downloads: 0
CorpDocs-Authorization (Proxy) To Vote Shares
Views: 215  |  Downloads: 5