Differences

This shows you the differences between two versions of the page.

--- b:head_first_statistics:geometric_distribution [2025/10/06 21:34] – hkimscil
+++ b:head_first_statistics:geometric_distribution [2025/10/06 21:39] (current) – [e.g.,] hkimscil
@@ Line 1: / Line 1: @@
-====== Geometric Distributions ======
+====== Geometric Distribution ======
-===== 정리 =====
+기하분포
-기하분포, 이항분포, 포아송분포
 \begin{align*}
 \text{Geometric Distribution:  } \;\;\; \text{X} & \thicksim Geo(p) \\
@@ Line 148: / Line 147: @@
 {{:b:head_first_statistics:pasted:20191031-005411.png}}
+====== Expected value ======
+X가 성공할 확률 p를 가진 Geometric distribution을 따른다  :: $X \sim \text{Geo}(p)$
+Reminding . . . [[:b/head_first_statistics/using_discrete_probability_distributions#expected_value_discrete_prob|Expected value in discrete probability]]
+$E(X) = \sum x*P(X=x)$
+| textbook  | x  | P(X = x)  | xP(X = x)  | xP(X ≤ x): \\  $E(X) = \sum (x*P(X=x))$  |
+| r code  | trial  | ''px <- q^(trial-1)*p''  | ''npx <- trial*(q^(trial-1))*p''  |  ''plex <- cumsum(trial*(q^(trial-1))*p)''  |
+|   |   | ''px''   | ''npx <- trial*px''   | ''plex <- cumsum(npx)''   |
+|   |   | x번째 (trial번째) \\ 성공할 확률  | x번째의 기대치 \\ (주사위 경우처럼)  | 그 x번째까지 성공할 \\ 확률에 대한 기대값  |
+  * 우리가 작업하고 있는 채드의 슬로프 타기 예가 얼른 이해가 안된다면 아래 workout의 예를 들어 본다.
+^ x  ^ p(x) px ^ npx.0       ^  npx = weighted \\ probability at \\ a given spot  ^ plex.0  ^  plex  ^   ^
+| 0  | 0.1     | 0 * 0.1   | 0.00   | 0.00  | 0.00  |   |
+| 1  | 0.15    | 1 * 0.15  | 0.15   | 0.00 + 0.15 | 0.15  |   |
+| 2  | 0.4     | 2 * 0.4   | 0.80   | 0.00 + 0.15 + 0.80  | 0.95  |   |
+| 3  | 0.25    | 3 * 0.25  | 0.75   | 0.00 + 0.15 + 0.80 + 0.75  | 1.7  |   |
+| 4  | 0.1     | 4 * 0.1   | 0.40   | 0.00 + 0.15 + 0.80 + 0.75 + 0.40  | 2.1  | = this is E(x)  |
+  * x 일주일에 내가 갈 운동횟수 (workout frequency, 0 to 4)
+  * px 각 횟수에 대한 probability
+  * npx weighted probability
+  * plex cumulative sum of npx (to find out the below last one)
+  * sum of npx = 2.1 = mean of all = expected value of x = E(x)
+  * https://www.youtube.com/watch?v=qafPcWNUiM8 참조
+<code>
+p <- .2
+q <- 1-p
+trial <- c(1:8)
+px <- q^(trial-1)*p
+px
+## npx <- trial*(q^(trial-1))*p
+## 위는 아래와 같음
+npx <- trial*px
+npx
+## plex <- cumsum(trial*(q^(trial-1))*p)
+## 위는 아래와 같음
+plex <- cumsum(npx)
+plex
+sumgeod <- data.frame(trial,px,npx,plex)
+round(sumgeod,3)
+</code>
+<code>
+> p <- .2
+> q <- 1-p
+> trial <- c(1,2,3,4,5,6,7,8)
+> px <- q^(trial-1)*p
+> px
+[1] 0.20000000 0.16000000 0.12800000 0.10240000 0.08192000 0.06553600 0.05242880 0.04194304
+> npx <- trial*(q^(trial-1))*p
+> npx
+[1] 0.2000000 0.3200000 0.3840000 0.4096000 0.4096000 0.3932160 0.3670016 0.3355443
+> plex <- cumsum(trial*(q^(trial-1))*p)
+> plex
+[1] 0.200000 0.520000 0.904000 1.313600 1.723200 2.116416 2.483418 2.818962
+> sumgeod <- data.frame(trial,px,npx,plex)
+> round(sumgeod,3)
+  trial    px   npx  plex
+     1 0.200 0.200 0.200
+     2 0.160 0.320 0.520
+     3 0.128 0.384 0.904
+     4 0.102 0.410 1.314
+     5 0.082 0.410 1.723
+     6 0.066 0.393 2.116
+     7 0.052 0.367 2.483
+     8 0.042 0.336 2.819
+>
+</code>
+  * 아래의 예는 위의 workout 예처럼 횟수가 0-4로 정해져 있지 않고 계속 진행됨 (0-무한대)
+  * 하지만 여기서는 100 까지로 한정 (1:100)
+  * 각 지점에서의 probability = geometric probability = q^(trial-1)*p = px
+  * 각 지점에서의 weighted prob = trial * px = npx
+  * 각 단계에서의 기대값을 구하기 위한 누적합계 cumsum(npx) = plex
+  * 아래 그림에서 plex는 각 단계의 probability density를 더해온 값을 말한다.
+  * 그림이 암시하는 것처럼 오른 쪽으로 한 없이 가면서 생기는 그래프의 용적은 기대값이 된다.
+| {{:b:head_first_statistics:pasted:20251002-072502.png?400}} |
+| {{:b:head_first_statistics:pasted:20251002-072514.png?400}} |
+| {{:b:head_first_statistics:pasted:20251002-072522.png?400}} |
+<code>
+p <- .2
+q <- 1-p
+trial <- c(1:100)
+px <- q^(trial-1)*p
+px
+npx <- trial*px
+npx
+## plex <- cumsum(trial*(q^(trial-1))*p)
+## 위는 아래와 같음
+plex <- cumsum(npx)
+plex
+sumgeod <- data.frame(trial,px,npx,plex)
+sumgeod
+plot(npx, type="l")
+plot(plex, type="l")
+</code>
+<code>
+>
+> p <- .2
+> q <- 1-p
+> trial <- c(1:100)
+> px <- q^(trial-1)*p
+> px
+  [1] 2.000000e-01 1.600000e-01 1.280000e-01 1.024000e-01
+  [5] 8.192000e-02 6.553600e-02 5.242880e-02 4.194304e-02
+  [9] 3.355443e-02 2.684355e-02 2.147484e-02 1.717987e-02
+ [13] 1.374390e-02 1.099512e-02 8.796093e-03 7.036874e-03
+ [17] 5.629500e-03 4.503600e-03 3.602880e-03 2.882304e-03
+ [21] 2.305843e-03 1.844674e-03 1.475740e-03 1.180592e-03
+ [25] 9.444733e-04 7.555786e-04 6.044629e-04 4.835703e-04
+ [29] 3.868563e-04 3.094850e-04 2.475880e-04 1.980704e-04
+ [33] 1.584563e-04 1.267651e-04 1.014120e-04 8.112964e-05
+ [37] 6.490371e-05 5.192297e-05 4.153837e-05 3.323070e-05
+ [41] 2.658456e-05 2.126765e-05 1.701412e-05 1.361129e-05
+ [45] 1.088904e-05 8.711229e-06 6.968983e-06 5.575186e-06
+ [49] 4.460149e-06 3.568119e-06 2.854495e-06 2.283596e-06
+ [53] 1.826877e-06 1.461502e-06 1.169201e-06 9.353610e-07
+ [57] 7.482888e-07 5.986311e-07 4.789049e-07 3.831239e-07
+ [61] 3.064991e-07 2.451993e-07 1.961594e-07 1.569275e-07
+ [65] 1.255420e-07 1.004336e-07 8.034690e-08 6.427752e-08
+ [69] 5.142202e-08 4.113761e-08 3.291009e-08 2.632807e-08
+ [73] 2.106246e-08 1.684997e-08 1.347997e-08 1.078398e-08
+ [77] 8.627183e-09 6.901746e-09 5.521397e-09 4.417118e-09
+ [81] 3.533694e-09 2.826955e-09 2.261564e-09 1.809251e-09
+ [85] 1.447401e-09 1.157921e-09 9.263367e-10 7.410694e-10
+ [89] 5.928555e-10 4.742844e-10 3.794275e-10 3.035420e-10
+ [93] 2.428336e-10 1.942669e-10 1.554135e-10 1.243308e-10
+ [97] 9.946465e-11 7.957172e-11 6.365737e-11 5.092590e-11
+> npx <- trial*px
+> npx
+  [1] 2.000000e-01 3.200000e-01 3.840000e-01 4.096000e-01
+  [5] 4.096000e-01 3.932160e-01 3.670016e-01 3.355443e-01
+  [9] 3.019899e-01 2.684355e-01 2.362232e-01 2.061584e-01
+ [13] 1.786706e-01 1.539316e-01 1.319414e-01 1.125900e-01
+ [17] 9.570149e-02 8.106479e-02 6.845471e-02 5.764608e-02
+ [21] 4.842270e-02 4.058284e-02 3.394201e-02 2.833420e-02
+ [25] 2.361183e-02 1.964504e-02 1.632050e-02 1.353997e-02
+ [29] 1.121883e-02 9.284550e-03 7.675228e-03 6.338253e-03
+ [33] 5.229059e-03 4.310012e-03 3.549422e-03 2.920667e-03
+ [37] 2.401437e-03 1.973073e-03 1.619997e-03 1.329228e-03
+ [41] 1.089967e-03 8.932412e-04 7.316071e-04 5.988970e-04
+ [45] 4.900066e-04 4.007165e-04 3.275422e-04 2.676089e-04
+ [49] 2.185473e-04 1.784060e-04 1.455793e-04 1.187470e-04
+ [53] 9.682448e-05 7.892109e-05 6.430607e-05 5.238022e-05
+ [57] 4.265246e-05 3.472060e-05 2.825539e-05 2.298743e-05
+ [61] 1.869645e-05 1.520236e-05 1.235804e-05 1.004336e-05
+ [65] 8.160232e-06 6.628619e-06 5.383242e-06 4.370871e-06
+ [69] 3.548119e-06 2.879633e-06 2.336616e-06 1.895621e-06
+ [73] 1.537559e-06 1.246898e-06 1.010998e-06 8.195824e-07
+ [77] 6.642931e-07 5.383362e-07 4.361904e-07 3.533694e-07
+ [81] 2.862292e-07 2.318103e-07 1.877098e-07 1.519771e-07
+ [85] 1.230291e-07 9.958120e-08 8.059129e-08 6.521410e-08
+ [89] 5.276414e-08 4.268560e-08 3.452790e-08 2.792587e-08
+ [93] 2.258353e-08 1.826109e-08 1.476428e-08 1.193576e-08
+ [97] 9.648071e-09 7.798028e-09 6.302080e-09 5.092590e-09
+> ## plex <- cumsum(trial*(q^(trial-1))*p)
+> ## 위는 아래와 같음
+> plex <- cumsum(npx)
+> plex
+  [1] 0.200000 0.520000 0.904000 1.313600 1.723200 2.116416 2.483418
+  [8] 2.818962 3.120952 3.389387 3.625610 3.831769 4.010440 4.164371
+ [15] 4.296313 4.408903 4.504604 4.585669 4.654124 4.711770 4.760192
+ [22] 4.800775 4.834717 4.863051 4.886663 4.906308 4.922629 4.936169
+ [29] 4.947388 4.956672 4.964347 4.970686 4.975915 4.980225 4.983774
+ [36] 4.986695 4.989096 4.991069 4.992689 4.994018 4.995108 4.996002
+ [43] 4.996733 4.997332 4.997822 4.998223 4.998550 4.998818 4.999037
+ [50] 4.999215 4.999361 4.999479 4.999576 4.999655 4.999719 4.999772
+ [57] 4.999814 4.999849 4.999877 4.999900 4.999919 4.999934 4.999947
+ [64] 4.999957 4.999965 4.999971 4.999977 4.999981 4.999985 4.999988
+ [71] 4.999990 4.999992 4.999993 4.999995 4.999996 4.999997 4.999997
+ [78] 4.999998 4.999998 4.999998 4.999999 4.999999 4.999999 4.999999
+ [85] 4.999999 5.000000 5.000000 5.000000 5.000000 5.000000 5.000000
+ [92] 5.000000 5.000000 5.000000 5.000000 5.000000 5.000000 5.000000
+ [99] 5.000000 5.000000
+> sumgeod <- data.frame(trial,px,npx,plex)
+> sumgeod
+    trial           px          npx     plex
+       1 2.000000e-01 2.000000e-01 0.200000
+       2 1.600000e-01 3.200000e-01 0.520000
+       3 1.280000e-01 3.840000e-01 0.904000
+       4 1.024000e-01 4.096000e-01 1.313600
+       5 8.192000e-02 4.096000e-01 1.723200
+       6 6.553600e-02 3.932160e-01 2.116416
+       7 5.242880e-02 3.670016e-01 2.483418
+       8 4.194304e-02 3.355443e-01 2.818962
+       9 3.355443e-02 3.019899e-01 3.120952
+     10 2.684355e-02 2.684355e-01 3.389387
+     11 2.147484e-02 2.362232e-01 3.625610
+     12 1.717987e-02 2.061584e-01 3.831769
+     13 1.374390e-02 1.786706e-01 4.010440
+     14 1.099512e-02 1.539316e-01 4.164371
+     15 8.796093e-03 1.319414e-01 4.296313
+     16 7.036874e-03 1.125900e-01 4.408903
+     17 5.629500e-03 9.570149e-02 4.504604
+     18 4.503600e-03 8.106479e-02 4.585669
+     19 3.602880e-03 6.845471e-02 4.654124
+     20 2.882304e-03 5.764608e-02 4.711770
+     21 2.305843e-03 4.842270e-02 4.760192
+     22 1.844674e-03 4.058284e-02 4.800775
+     23 1.475740e-03 3.394201e-02 4.834717
+     24 1.180592e-03 2.833420e-02 4.863051
+     25 9.444733e-04 2.361183e-02 4.886663
+     26 7.555786e-04 1.964504e-02 4.906308
+     27 6.044629e-04 1.632050e-02 4.922629
+     28 4.835703e-04 1.353997e-02 4.936169
+     29 3.868563e-04 1.121883e-02 4.947388
+     30 3.094850e-04 9.284550e-03 4.956672
+     31 2.475880e-04 7.675228e-03 4.964347
+     32 1.980704e-04 6.338253e-03 4.970686
+     33 1.584563e-04 5.229059e-03 4.975915
+     34 1.267651e-04 4.310012e-03 4.980225
+     35 1.014120e-04 3.549422e-03 4.983774
+     36 8.112964e-05 2.920667e-03 4.986695
+     37 6.490371e-05 2.401437e-03 4.989096
+     38 5.192297e-05 1.973073e-03 4.991069
+     39 4.153837e-05 1.619997e-03 4.992689
+     40 3.323070e-05 1.329228e-03 4.994018
+     41 2.658456e-05 1.089967e-03 4.995108
+     42 2.126765e-05 8.932412e-04 4.996002
+     43 1.701412e-05 7.316071e-04 4.996733
+     44 1.361129e-05 5.988970e-04 4.997332
+     45 1.088904e-05 4.900066e-04 4.997822
+     46 8.711229e-06 4.007165e-04 4.998223
+     47 6.968983e-06 3.275422e-04 4.998550
+     48 5.575186e-06 2.676089e-04 4.998818
+     49 4.460149e-06 2.185473e-04 4.999037
+     50 3.568119e-06 1.784060e-04 4.999215
+     51 2.854495e-06 1.455793e-04 4.999361
+     52 2.283596e-06 1.187470e-04 4.999479
+     53 1.826877e-06 9.682448e-05 4.999576
+     54 1.461502e-06 7.892109e-05 4.999655
+     55 1.169201e-06 6.430607e-05 4.999719
+     56 9.353610e-07 5.238022e-05 4.999772
+     57 7.482888e-07 4.265246e-05 4.999814
+     58 5.986311e-07 3.472060e-05 4.999849
+     59 4.789049e-07 2.825539e-05 4.999877
+     60 3.831239e-07 2.298743e-05 4.999900
+     61 3.064991e-07 1.869645e-05 4.999919
+     62 2.451993e-07 1.520236e-05 4.999934
+     63 1.961594e-07 1.235804e-05 4.999947
+     64 1.569275e-07 1.004336e-05 4.999957
+     65 1.255420e-07 8.160232e-06 4.999965
+     66 1.004336e-07 6.628619e-06 4.999971
+     67 8.034690e-08 5.383242e-06 4.999977
+     68 6.427752e-08 4.370871e-06 4.999981
+     69 5.142202e-08 3.548119e-06 4.999985
+     70 4.113761e-08 2.879633e-06 4.999988
+     71 3.291009e-08 2.336616e-06 4.999990
+     72 2.632807e-08 1.895621e-06 4.999992
+     73 2.106246e-08 1.537559e-06 4.999993
+     74 1.684997e-08 1.246898e-06 4.999995
+     75 1.347997e-08 1.010998e-06 4.999996
+     76 1.078398e-08 8.195824e-07 4.999997
+     77 8.627183e-09 6.642931e-07 4.999997
+     78 6.901746e-09 5.383362e-07 4.999998
+     79 5.521397e-09 4.361904e-07 4.999998
+     80 4.417118e-09 3.533694e-07 4.999998
+     81 3.533694e-09 2.862292e-07 4.999999
+     82 2.826955e-09 2.318103e-07 4.999999
+     83 2.261564e-09 1.877098e-07 4.999999
+     84 1.809251e-09 1.519771e-07 4.999999
+     85 1.447401e-09 1.230291e-07 4.999999
+     86 1.157921e-09 9.958120e-08 5.000000 ###########
+     87 9.263367e-10 8.059129e-08 5.000000
+     88 7.410694e-10 6.521410e-08 5.000000
+     89 5.928555e-10 5.276414e-08 5.000000
+     90 4.742844e-10 4.268560e-08 5.000000
+     91 3.794275e-10 3.452790e-08 5.000000
+     92 3.035420e-10 2.792587e-08 5.000000
+     93 2.428336e-10 2.258353e-08 5.000000
+     94 1.942669e-10 1.826109e-08 5.000000
+     95 1.554135e-10 1.476428e-08 5.000000
+     96 1.243308e-10 1.193576e-08 5.000000
+     97 9.946465e-11 9.648071e-09 5.000000
+     98 7.957172e-11 7.798028e-09 5.000000
+     99 6.365737e-11 6.302080e-09 5.000000
+   100 5.092590e-11 5.092590e-09 5.000000
+> plot(npx, type="l")
+> plot(plex, type="l")
+</code>
+  * 기댓값이 86번째 부터는 더이상 늘지 않고
+  * 계산된 값을 보면 5로 수렴한다.
+  * workout 예처럼 다섯가지의 순서가 있는 것이 아니라서
+  * 평균을 어떻게 나오나 보기 위해서 100까지 해 봤지만
+  * 86번째 이후에는 평균값이 더 늘지 않는다 (5에서)
+  * 따라서 위의 geometric distribution에서의 기대값은 5이다.
+{{:b:head_first_statistics:pasted:20191031-010726.png}}
+{{:b:head_first_statistics:pasted:20191031-010811.png}}
+  * 그런데 이 기대값은 아래처럼 구할 수 있다.
+  * 위에서 $X \sim \text{Geo}(p)$ 일때, 기대값은 $E(X) = \dfrac{1}{p}$
+  * 아래는 그 증명이다.
+====== Proof of mean and variance of geometric distribution ======
+$(4)$, $(5)$에 대한 증명은 [[:Mean and Variance of Geometric Distribution]]
+===== e.g., =====
+<WRAP box>
+The probability that another snowboarder will make it down the slope without falling over is 0.4. Your job is to play like you’re the snowboarder and work out the following probabilities for your slope success.
+  - The probability that you will be successful on your second attempt, while failing on your first.
+  - The probability that you will be successful in 4 attempts or fewer.
+  - The probability that you will need more than 4 attempts to be successful.
+  - The number of attempts you expect you’ll need to make before being successful.
+  - The variance of the number of attempts.
+</WRAP>
+  - $P(X = 2) = p * q^{2-1}$
+  - $P(X \le 4) = 1 - q^{4}$
+  - $P(X > 4) = q^{4}$
+  - $E(X) = \displaystyle \frac{1}{p}$
+  - $Var(X) = \displaystyle \frac{q}{p^{2}}$