-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathhagen_ranking.R
246 lines (230 loc) · 10.3 KB
/
hagen_ranking.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
# Copyright 2015 IBM Corp. All Rights Reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
# ***** AUTHOR *****
# Irving A Duran/Dubuque/IBM ([email protected])
# ***** CREATED *****
# Aug 25, 2015
# ***** LAST DATE MODIFIED *****
# Sept 16, 2015
# ***** PROGRAM NAME *****
# Hagen Ranking
# ***** VERSION *****
# 1.3
# ***** BUILD *****
# 0
# ***** PROGRAM DESCRIPTION *****
# Percentile ranking based on book "Introductory Statistics: Concepts, Models, and Applications by David W. Stockburger"
# R Built > 2.15.0
# ********************************************************************************
# Any code/syntax that is comment out means that might still be in development
# OR would allow you to perform some validation on your dataset.
# ********************************************************************************
# install packages if they don't exist
if (!require("plyr")){
install.packages("plyr", dependencies = TRUE)
library("plyr")
}
completeFun <- function(data, desiredCols) {
completeVec <- complete.cases(data[, desiredCols])
return(data[completeVec, ])
}
#population/parameters formulas
#population mean = μ = ( Σ Xi ) / N
pop.mean <- function(x){ sum(x) / length(x) }
#population variance = σ2 = Σ ( Xi - μ )2 / N
pop.var <- function(x){ sum((x-pop.mean(x))^2)/ length(x) }
#population standard deviation = σ = sqrt [ Σ ( Xi - μ )2 / N ]
pop.sd <- function(x){ sqrt(pop.var(x)) }
#standardized score (z-score)= Z = (X - μ) / σ
z.s <- function(x, mean, sd){ (x - mean) / sd }
# store variables from modeler
df <- modelerData
df$idx <- as.numeric(rownames(df))
s <- "%%rankingfield%%" #transfer field names selected to modeler
namev <- strsplit(s, ", ") #split variables based on commas
# check if z-score check-box is checked
if ("%%zscorerank%%"=="TRUE"){
# ********************************************************************************
# Working with z-scores instead of actual raw score to be ranked
# ********************************************************************************
for (j in 1:length(namev[[1]])){
z.score <- 0 #sanitize
z.scoreidx <- 0 #sanitize
df.zs <- completeFun(df[, c("idx", namev[[1]][j])], namev[[1]][j])
df.zs <- data.frame(df.zs, row.names = seq_along(df.zs$idx)) #reset table index to avoid issues with counts
z.score <- z.s(df.zs[, namev[[1]][j]],
pop.mean(df.zs[, namev[[1]][j]]),
pop.sd(df.zs[, namev[[1]][j]])) #get z-scores based on function
for (i in 1:length(df.zs[, namev[[1]][j]])){
z.scoreidx[i] <- df.zs[, "idx"][i]
}
df.r <- data.frame(z.score, z.scoreidx)
df.r <- rename(df.r, c("z.score"=c(paste(namev[[1]],".zscore", sep="")[j]),
"z.scoreidx"="idx")) #rename column
df <- merge(df, df.r, by="idx", all.x=TRUE) #merge data by x
#clean up after loop
if(j==length(namev[[1]])){
df <- df[with(df, order(idx)), ] # sort by initial index
#df <- df[ , -which(names(df) %in% c("idx"))] #exclude idx column
rm(z.score, z.scoreidx, df.zs, df.r)
# replace name prior of ranking using z-score
#namev <- paste(namev[[1]],".zscore", sep=""); namev <- list(namev);
}
}
# ********************************************************************************
# Start ranking process
# ********************************************************************************
# get frequency count
freq <- apply(df[namev[[1]]], 2, count)
# provide names to variables from list
lst <- lapply(seq(namev[[1]]),
function(j){
y <- data.frame(freq[[j]])
names(y) <- c(namev[[1]][j], paste(namev[[1]],".freq", sep="")[j])
return(y)
}
)
# provide names to variables from list
names(lst) <- namev[[1]]
# loop through list and perform merge
for (i in 1:length(namev[[1]])){
t <- as.data.frame(lst[i])
t <- t[complete.cases(t), ] #skip records with NA/Blank/Null
names(t) <- c(namev[[1]][i], paste(namev[[1]],".freq", sep="")[i])
df <- merge(df, t, by=namev[[1]][i], all=TRUE) #merge data by x
}
# ranking formula for Hagen -> source=http://www.psychstat.missouristate.edu/introbook/sbk14.htm
# PR=Probability Ranking
# Fb=Frequency below -> is the frequency below; the number of scores which are less than the score value of the percentile rank
# Fw=Frequency within -> is the frequency within; the number of scores which have the same value as the score value of the percentile rank
# N=Number of scores
# PR = ((Fb + (1/2 * Fw)) / N) * 100
# loop through name list and then rank them according to Hagen ranking formula for each data frame
rank <- 0
rankidx <- 0
for (j in 1:length(namev[[1]])){
rank <- 0 #sanitize
rankidx <- 0 #sanitize
df.t <- completeFun(df[, c("idx", namev[[1]][j], paste(namev[[1]][j],".freq", sep=""))], namev[[1]][j])
df.t <- data.frame(df.t, row.names = seq_along(df.t$idx)) #reset table index to avoid issues with counts
for (i in 1:length(df.t[, namev[[1]][j]])){
fw <- 1 #sanitize
fb <- length(which(df.t[, namev[[1]][j]] < df.t[, namev[[1]][j]][i]))
if((df.t[, paste(namev[[1]][j],".freq", sep="")][i] > 1)=="TRUE"){
fw <- df.t[, paste(namev[[1]][j],".freq", sep="")][i] #store freq count
rank[i] <- ((fb + ((1/2)*fw)) / length(df.t[, namev[[1]][j]]))
rankidx[i] <- df.t[, "idx"][i]
}
else{
rank[i] <- ((fb + ((1/2)*fw)) / length(df.t[, namev[[1]][j]]))
rankidx[i] <- df.t[, "idx"][i]
}
}
df.r <- data.frame(rank, rankidx)
df.r <- rename(df.r, c("rank"=c(paste(namev[[1]],".rank", sep="")[j]),
"rankidx"="idx")) #rename column
df <- merge(df, df.r, by="idx", all.x=TRUE) #merge data by x
df <- df[ , -which(names(df) %in% c(paste(namev[[1]],".freq", sep="")[j]))] #exclude freq columns
#clean up after loop
if(j==length(namev[[1]])){
df <- df[with(df, order(idx)), ] # sort by initial index
df <- df[ , -which(names(df) %in% c("idx"))] #exclude idx column
rm(rank, rankidx, df.t, df.r)
}
}
#df;
# store created data back into modeler
for (j in 1:length(namev[[1]])){
modelerData <- cbind(modelerData, df[paste(namev[[1]],".rank", sep="")[j]])
var1 <- c(fieldName=paste(namev[[1]],".rank", sep="")[j], fieldLabel="", fieldStorage="real", fieldMeasure="", fieldFormat="", fieldRole="")
modelerDataModel <- data.frame(modelerDataModel, var1)
modelerData <- cbind(modelerData, df[paste(namev[[1]],".zscore", sep="")[j]])
var2 <- c(fieldName=paste(namev[[1]],".zscore", sep="")[j], fieldLabel="", fieldStorage="real", fieldMeasure="", fieldFormat="", fieldRole="")
modelerDataModel <- data.frame(modelerDataModel, var2)
}
}
if ("%%zscorerank%%"=="FALSE"){
# ********************************************************************************
# Start ranking process
# ********************************************************************************
# get frequency count
freq <- apply(df[namev[[1]]], 2, count)
# provide names to variables from list
lst <- lapply(seq(namev[[1]]),
function(j){
y <- data.frame(freq[[j]])
names(y) <- c(namev[[1]][j], paste(namev[[1]],".freq", sep="")[j])
return(y)
}
)
# provide names to variables from list
names(lst) <- namev[[1]]
# loop through list and perform merge
for (i in 1:length(namev[[1]])){
t <- as.data.frame(lst[i])
t <- t[complete.cases(t), ] #skip records with NA/Blank/Null
names(t) <- c(namev[[1]][i], paste(namev[[1]],".freq", sep="")[i])
df <- merge(df, t, by=namev[[1]][i], all=TRUE) #merge data by x
}
# ranking formula for Hagen -> source=http://www.psychstat.missouristate.edu/introbook/sbk14.htm
# PR=Probability Ranking
# Fb=Frequency below -> is the frequency below; the number of scores which are less than the score value of the percentile rank
# Fw=Frequency within -> is the frequency within; the number of scores which have the same value as the score value of the percentile rank
# N=Number of scores
# PR = ((Fb + (1/2 * Fw)) / N) * 100
# loop through name list and then rank them according to Hagen ranking formula for each data frame
rank <- 0
rankidx <- 0
for (j in 1:length(namev[[1]])){
rank <- 0 #sanitize
rankidx <- 0 #sanitize
df.t <- completeFun(df[, c("idx", namev[[1]][j], paste(namev[[1]][j],".freq", sep=""))], namev[[1]][j])
df.t <- data.frame(df.t, row.names = seq_along(df.t$idx)) #reset table index to avoid issues with counts
for (i in 1:length(df.t[, namev[[1]][j]])){
fw <- 1 #sanitize
fb <- length(which(df.t[, namev[[1]][j]] < df.t[, namev[[1]][j]][i]))
if((df.t[, paste(namev[[1]][j],".freq", sep="")][i] > 1)=="TRUE"){
fw <- df.t[, paste(namev[[1]][j],".freq", sep="")][i] #store freq count
rank[i] <- ((fb + ((1/2)*fw)) / length(df.t[, namev[[1]][j]]))
rankidx[i] <- df.t[, "idx"][i]
}
else{
rank[i] <- ((fb + ((1/2)*fw)) / length(df.t[, namev[[1]][j]]))
rankidx[i] <- df.t[, "idx"][i]
}
}
df.r <- data.frame(rank, rankidx, qnorm(rank))
df.r <- rename(df.r, c("rank"=c(paste(namev[[1]],".rank", sep="")[j]),
"rankidx"="idx",
"qnorm.rank."=c(paste(namev[[1]],".zscore", sep="")[j]))) #rename column
df <- merge(df, df.r, by="idx", all.x=TRUE) #merge data by x
df <- df[ , -which(names(df) %in% c(paste(namev[[1]],".freq", sep="")[j]))] #exclude freq columns
#clean up after loop
if(j==length(namev[[1]])){
df <- df[with(df, order(idx)), ] # sort by initial index
df <- df[ , -which(names(df) %in% c("idx"))] #exclude idx column
rm(rank, rankidx, df.t, df.r)
}
}
#df;
# store created data back into modeler
for (j in 1:length(namev[[1]])){
modelerData <- cbind(modelerData, df[paste(namev[[1]],".rank", sep="")[j]])
var1 <- c(fieldName=paste(namev[[1]],".rank", sep="")[j], fieldLabel="", fieldStorage="real", fieldMeasure="", fieldFormat="", fieldRole="")
modelerDataModel <- data.frame(modelerDataModel, var1)
modelerData <- cbind(modelerData, df[paste(namev[[1]],".zscore", sep="")[j]])
var2 <- c(fieldName=paste(namev[[1]],".zscore", sep="")[j], fieldLabel="", fieldStorage="real", fieldMeasure="", fieldFormat="", fieldRole="")
modelerDataModel <- data.frame(modelerDataModel, var2)
}
}