74:
53:
22:
146:
I noticed an oddity in the definition of the circular variance, which probably has historical reasons: While it is not generally true that the estimator of the circular standard deviation squared yields the circular standard deviation, I would have expected this equality to hold for very small
434:
The current title of this article is directional statistics, which is a synonym of circular statistics. A quick search on Google
Scholar and Google Books will show that the latter term, "circular statistics," is more widely used than "directional statistics". Here are the results.
342:
Consider the same example but rotated further to the left so that the three angles are 330, 340 and 350 degrees. Taking the modulus 360 of the sum (1020) results in 2 remainder 280, which divided by 3 is clearly not 340.
124:
260:
147:
values of the circular standard deviation, which recovers the linear case. However, there is a factor of 2 missing! You can test this numerically yourself with the following line of code
374:
I've added the wrapped normal distribution as an example of the way a wrapped distribution can be made from the pdf of another distribution. I'm also redirecting the nonexistent
198:
513:
114:
324:
The final step of the formula given for the example is unclear; apparently it is written for result in radians. Wouldn't it be better written this way:
518:
327:
Mean angle = arctangent (mean sine / mean cosine). If mean cosine < 0, add 180 degrees to the result. If mean sine <0 and mean cosine : -->
360:
Agreed - I've removed it. Also, this article is about
Directional statistics itself, not about various methods to calculate a circular average.
90:
508:
81:
58:
346:
As the remainder (in a modulus 360 operation) can only range from 0 to 360, the 'average' can only range from 0 to 360/n.
276:
33:
375:
203:
272:
387:
469:
39:
73:
52:
487:
451:
21:
352:
305:
89:
on
Knowledge. If you would like to participate, please visit the project page, where you can join
309:
401:
379:
465:
295:
N ( \mu ,\sigma^2) round the circle gives the wrapped normal distribution W N ( \mu , rho )
483:
447:
446:
Thus, I would like to propose renaming this article to
Circular statistics. Any thoughts?
420:
174:
439:"Circular statistics": 9,670 results on Google Scholar and 811 results on Google Books
502:
298:
with \rho = exp{ -\sigma^2/2) as in (3.5.63), so that 1 - \nu =\exp(- \sigma^2 / 2 ).
397:
361:
491:
473:
455:
424:
405:
391:
364:
355:
339:
The modulus method only works in few specific cases (such as the example given).
313:
280:
442:"Directional statistics": 4,060 on Google Scholar and 766 results on Google Books
415:
I started a new
Goodness of fit section - in case it should be put elsewhere. -
416:
86:
292:"The motivation for this definition is that wrapping the normal distribution
464:: circular is 2D, a special case of spherical statistics or 3D directions.
349:
Unless I'm misunderstanding the procedure, I would suggest taking it out.
15:
156:
std_circ=sqrt(-2*log(abs(mean(exp(1i*sigma*randn(1e5,1))))))
165:
I therefore added the following sentence to the article:
301:
For small \nu, (3.4.14) reduces to \sigma = \sqrt(2\nu).
206:
177:
85:, a collaborative effort to improve the coverage of
254:
192:
161:var_circ=1-abs(mean(exp(1i*sigma*randn(1e5,1))))
378:to this page. I hope this is considered useful!
286:Thank for adding this. It also got me confused.
8:
255:{\displaystyle S(z)^{2}=2\mathrm {Var} (z)}
47:
232:
220:
205:
176:
142:Circular standard deviation | Factor of 2
268:Does anybody know a reference for this?
49:
19:
411:Should the Rayleigh test be mentioned
328:= 0, add 360 degrees to the result.
7:
79:This article is within the scope of
514:High-importance Statistics articles
38:It is of interest to the following
239:
236:
233:
14:
289:The Mardia & Jupp book says:
99:Knowledge:WikiProject Statistics
72:
51:
20:
519:WikiProject Statistics articles
119:This article has been rated as
102:Template:WikiProject Statistics
249:
243:
217:
210:
187:
181:
1:
406:07:07, 12 November 2008 (UTC)
392:22:18, 11 November 2008 (UTC)
93:and see a list of open tasks.
509:C-Class Statistics articles
492:03:15, 28 August 2015 (UTC)
474:03:10, 28 August 2015 (UTC)
456:02:48, 28 August 2015 (UTC)
430:Propose renaming of article
376:wrapped normal distribution
365:09:42, 20 August 2007 (UTC)
356:09:28, 20 August 2007 (UTC)
281:20:39, 25 August 2014 (UTC)
535:
425:14:31, 6 April 2015 (UTC)
118:
67:
46:
314:18:54, 6 May 2024 (UTC)
256:
194:
82:WikiProject Statistics
28:This article is rated
257:
195:
204:
193:{\displaystyle S(z)}
175:
171:Note that for small
396:Good work, thanks.
105:Statistics articles
273:Benjamin.friedrich
252:
190:
34:content assessment
139:
138:
135:
134:
131:
130:
526:
261:
259:
258:
253:
242:
225:
224:
199:
197:
196:
191:
162:
157:
152:
125:importance scale
107:
106:
103:
100:
97:
76:
69:
68:
63:
55:
48:
31:
25:
24:
16:
534:
533:
529:
528:
527:
525:
524:
523:
499:
498:
432:
413:
372:
337:
322:
216:
202:
201:
173:
172:
160:
155:
150:
144:
121:High-importance
104:
101:
98:
95:
94:
62:High‑importance
61:
32:on Knowledge's
29:
12:
11:
5:
532:
530:
522:
521:
516:
511:
501:
500:
497:
496:
495:
494:
477:
476:
444:
443:
440:
431:
428:
412:
409:
371:
370:wrapped normal
368:
336:
335:Modulus method
333:
321:
318:
317:
316:
302:
299:
296:
293:
290:
287:
266:
265:
264:
263:
251:
248:
245:
241:
238:
235:
231:
228:
223:
219:
215:
212:
209:
189:
186:
183:
180:
143:
140:
137:
136:
133:
132:
129:
128:
117:
111:
110:
108:
91:the discussion
77:
65:
64:
56:
44:
43:
37:
26:
13:
10:
9:
6:
4:
3:
2:
531:
520:
517:
515:
512:
510:
507:
506:
504:
493:
489:
485:
482:Point taken.
481:
480:
479:
478:
475:
471:
467:
463:
460:
459:
458:
457:
453:
449:
441:
438:
437:
436:
429:
427:
426:
422:
418:
410:
408:
407:
403:
399:
394:
393:
389:
385:
383:
377:
369:
367:
366:
363:
358:
357:
354:
350:
347:
344:
340:
334:
332:
329:
325:
319:
315:
311:
307:
303:
300:
297:
294:
291:
288:
285:
284:
283:
282:
278:
274:
269:
246:
229:
226:
221:
213:
207:
184:
178:
170:
169:
168:
167:
166:
163:
158:
153:
148:
141:
126:
122:
116:
113:
112:
109:
92:
88:
84:
83:
78:
75:
71:
70:
66:
60:
57:
54:
50:
45:
41:
35:
27:
23:
18:
17:
461:
445:
433:
414:
395:
381:
373:
359:
351:
348:
345:
341:
338:
330:
326:
323:
270:
267:
164:
159:
154:
149:
145:
120:
80:
40:WikiProjects
466:fgnievinski
151:sigma=0.1
503:Categories
484:danielkueh
448:danielkueh
200:, we have
96:Statistics
87:statistics
59:Statistics
353:Fink3412
306:Cosine12
398:Tomixdf
362:Tomixdf
123:on the
30:C-class
462:Oppose
384:enough
36:scale.
417:Rod57
488:talk
470:talk
452:talk
421:talk
402:talk
388:talk
331:Jim
310:talk
277:talk
271:Ben
115:High
382:far
380:dig
505::
490:)
472:)
454:)
423:)
404:)
390:)
312:)
304:"
279:)
486:(
468:(
450:(
419:(
400:(
386:(
320:?
308:(
275:(
262:.
250:)
247:z
244:(
240:r
237:a
234:V
230:2
227:=
222:2
218:)
214:z
211:(
208:S
188:)
185:z
182:(
179:S
127:.
42::
Text is available under the Creative Commons Attribution-ShareAlike License. Additional terms may apply.