0% found this document useful (0 votes)
14 views4 pages

Week 4 Thursday

Uploaded by

Mihir
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
14 views4 pages

Week 4 Thursday

Uploaded by

Mihir
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
You are on page 1/ 4
Data Analysis Section 1.6 1.6 Quantile-quantile (q-q) plots 1.6.1 Quantiles Quantiles are percentage points of a probability distribution. For example, the median is the 0.50 quantile of the distribution. And the upper quartile gy is the 0.75 quantile of the distribution, EXAMPLE 1.8. The 0.25 or 25% quantile or 25*# percentile (ie. the first quartile) of the standard Normal distribution is 0.6745, because if Z ~ N(0, 1) then Pr(Z < -0.6745) = 0.25 You can ee this from we “AB, OF from any pre set of statistical tables. Cee teet quantile 13 te tale og sb P( 2X) We attempt to define quantiles for data (rather than for a probability distribution) in a similar way. EXAMPLE 1.9. Find the quantiles of the following data (shown after sorting) 3,8 6.1 8.1 8.2 9.4/10.3 11.4 11 7| 12.4 12.9 i Solution, The smallest observation, x1 = 3.8, is the first of 10 in size, so we might think that it should be the 1/10 = 0.10 quantile. But this would make 5 = 9.4 the 0. Pa quantile, or the median. We know this is false, as the median is no. Aates pots (x5 + 26)/2 = (9-4 + 10.3)/2 = 9.85. ‘A neat way to overcome this is withthe following definition. |” For any set of data, the ith smallest value is the (i ~ 0.5)/n quantile. ‘Thus a, = 3.8 is the (1 — 0.8)/10 = 0.05 quantile, 2» = 6.1 is the (2 - 0.5)/10 = 0.15 quantile, ete. If we draw a graph of the quantiles versus the « values we get gregh, + c&malahre : & dishrbuwhon i Sun cho Aatw set ZPEM2302 Mathematical Tools for Science Let X be a ran dom variable with mean j1 and standard deviation 7. Show that the random. variable y=*X=# o has mean 0 and standard deviation 1 eye S (ZAP OL ee tale ale 2 2 £ with polabi P(X =x) Sy PLe ex) ~ LS pm tery Fe oo ees een Vi) = $ 9° PO=y) - EO) §iN@ “3 (sere 5 " We X= A S 2S HO P(X =a) x 6* FALE Pe) “tals Plt) LES Pike) e K(ge ht P(X= 2a) - ust ‘g) Lu a 6 £1 ¢ftea)- Pp) G Ce | ZPEM2302 Mathematical Tools for Science Let X ~ Binom(n,p) and suppose n is large in terms of p. Why can we approximate X by a normal distribution? woh P(x=k) ((%)p™U-pY Lek ¥~ NU 6), Ue Y bos pg. Ply=y) = pleas “ey imme © AQueshe 2 Rew ; Cor we appror inate fy te . (Le (1 f) b a corboves func hen * g (hy ae In- -h)! bey (KL) = By at > bah ~ log (nk): lay a! = lage lg a rset gr re,

You might also like