Editing 2236: Is it Christmas?
Warning: You are not logged in. Your IP address will be publicly visible if you make any edits. If you log in or create an account, your edits will be attributed to your username, along with other benefits.
The edit can be undone.
Please check the comparison below to verify that this is what you want to do, and then save the changes below to finish undoing the edit.
Latest revision | Your text | ||
Line 8: | Line 8: | ||
==Explanation== | ==Explanation== | ||
− | + | {{incomplete|Created by a BOT (99.73% accurate). Please mention here why this explanation isn't complete. Do NOT delete this tag too soon.}} | |
+ | Here Randall has made a comic which nearly always correctly tells if it is {{w|Christmas}}. Christmas is a holiday celebrated in the Western world on December 25 of each year. This also works for any annual event lasting only one day. | ||
− | + | Randall lists a rounded calculation of 99.73%. If we calculate without leap years, 364/365 shows 99.726027397% accuracy. Calculating with leap years gives 364.2425/365.2425: approximately 99.726209299%. Or running the calculation on just a leap year gives 365/366 (99.7267759562%). Because all three numbers round up to 99.73%, the listed percentage is correct all three ways. | |
− | |||
− | |||
− | |||
− | |||
Although Randall's claim on {{w|Accuracy and precision#In binary classification|accuracy}} is true, accuracy alone doesn't make a predictive device useful. In this case, the page {{w|False positives and false_negatives#false negative rate|miss rate}} or false negative rate, that is, the percent of positive condition days (it's Christmas) that are predicted by the comic not to be Christmas, is 100%. In other words, it misses all actual events of Christmas. | Although Randall's claim on {{w|Accuracy and precision#In binary classification|accuracy}} is true, accuracy alone doesn't make a predictive device useful. In this case, the page {{w|False positives and false_negatives#false negative rate|miss rate}} or false negative rate, that is, the percent of positive condition days (it's Christmas) that are predicted by the comic not to be Christmas, is 100%. In other words, it misses all actual events of Christmas. | ||
− | When building a model for rare events, a common mistake is to ignore the implicit cost function built into the standard prediction accuracy validity statistic for binary events. Prediction accuracy (# correct guesses/total guesses) assumes that false positives and false negatives are equally bad. Given the implicit cost function of this performance statistic, the best-performing model is commonly a persistence forecast model-- | + | When building a model for rare events, a common mistake is to ignore the implicit cost function built into the standard prediction accuracy validity statistic for binary events. Prediction accuracy (# correct guesses/total guesses) assumes that false positives and false negatives are equally bad. Given the implicit cost function of this performance statistic, the best-performing model is commonly a persistence forecast model--ie, the optimal prediction model returns the most common value whatever the model inputs are. It's probably a better choice to optimize a model using a performance statistic which relies on a cost function that penalizes missing correct prediction of rare events more than it penalizes missing correct prediction of common events. |
In fact, in most settings where a single outcome is a lot more common than any other one, predicting always that most common outcome would yield very high accuracy without any usefulness. It isn't hard to find examples even more accurate than Randall's: | In fact, in most settings where a single outcome is a lot more common than any other one, predicting always that most common outcome would yield very high accuracy without any usefulness. It isn't hard to find examples even more accurate than Randall's: | ||
Line 24: | Line 21: | ||
* A useless test for AIDS giving always negative results would have an accuracy about 99.95% when applied to a random human, and even more if used in countries with low prevalence of AIDS. | * A useless test for AIDS giving always negative results would have an accuracy about 99.95% when applied to a random human, and even more if used in countries with low prevalence of AIDS. | ||
* A website saying "You are not the cartoonist Randall Munroe" would be right for 99.9999999857% of humans. | * A website saying "You are not the cartoonist Randall Munroe" would be right for 99.9999999857% of humans. | ||
− | * | + | * A stopped watch is accurate twice a day while a running watch is almost never accurate (and oddly, is more accurate the faster/slower it runs). |
− | + | https://isitchristmas.com/ is a website that looks similar to the comic, with a '''NO''' printed if it is not Christmas, and a '''YES''' if it is Christmas. This website does a check on the computer's current date, and updates accordingly if it is indeed Christmas. Randall's comic doesn't do any of this, but as stated, is still correct most of the time. In addition, isitchristmas.com gives the answer in the language of your region (i.e. a visitor from Canada will give the answer in English and French to account for Canada's bilingularity); the strip only gives a fixed answer in English. | |
− | + | The title text is a "proof" that his service works. He claims to have tested this on 30 different days and confirmed that NO is the correct result. Any date except Christmas would result in a correct result. | |
− | + | This might be a reference to the phrase [https://knowyourphrase.com/even-a-broken-clock-is-right-twice A broken clock is right twice a day] | |
==Transcript== | ==Transcript== | ||
+ | :[A large square white panel with one large word in the middle, plus a footnote:] | ||
:'''<big><big><big>No*</big></big></big>''' | :'''<big><big><big>No*</big></big></big>''' | ||
:<nowiki>*</nowiki>99.73% accurate | :<nowiki>*</nowiki>99.73% accurate | ||
Line 38: | Line 36: | ||
:[Caption below the panel:] | :[Caption below the panel:] | ||
:xkcd.com presents a new "Is It Christmas" service to compete with isitchristmas.com | :xkcd.com presents a new "Is It Christmas" service to compete with isitchristmas.com | ||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
{{comic discussion}} | {{comic discussion}} | ||
[[Category:Christmas]] | [[Category:Christmas]] | ||
− |