A Complete Guide to Area Under Curve (AUC)

Deepanshu Bhalla 23 Comments , ,

This tutorial explains the various methods to calculate the AUC (Area under the ROC Curve) mathematically as well as the steps to implement it in Python, R and SAS.

What is Area under Curve?

Area under Curve (AUC) or Receiver operating characteristic (ROC) curve is used to evaluate the performance of a binary classification model. It measures discrimination power of a predictive classification model. In simple words, it checks how well model is able to distinguish between events and non-events.

Example : Suppose you are building a predictive model for bank to identify customers who are likely to purchase a credit card. In this case, purchase of credit card is event (or desired outcome) and non-purchase of credit card is non-event.

ROC Curve

AUC or ROC curve is a plot of the proportion of true positives (events correctly predicted to be events) versus the proportion of false positives (non-events wrongly predicted to be events) at different probability cutoffs. True Positive Rate is also called Sensitivity. False Positive Rate is also called (1-Specificity).

Higher the AUC score, better the model. Diagonal line represents random classification model. It is equivalent to prediction by tossing a coin. All points along the diagonal line say same true positive and false positive rate.

How to Generate ROC Curve?

To generate ROC curve, we calculate Sensitivity and (1-Specificity) at all possible cutoffs and then we plot them. Cut-off represents minimum threshold above that predicted probability would be classified as 'event'.

Let's say cutoff is 0.5. In the case of propensity to purchase a credit card, customers with a predicted probability greater than or equal to 0.5 would be classified as potential buyers.

Cut-off Sensitivity Specificity 1-specificity
0 1 0 1
0.01 0.979 0.081 0.919
0.02 0.938 0.158 0.842
….
….
….
0.99 0.02 0.996 0.004
1 0 1 0


Calculating AUC using Concordance and Tied Percent

  1. Calculate the predicted probability in logistic regression or any other binary classification technique.
  2. Divide the data into two datasets. One dataset contains observations having actual value of dependent variable with value 1 (i.e. event) and corresponding predicted probability values. And the other dataset contains observations having actual value of dependent variable 0 (non-event) against their predicted probability scores.
  3. Compare each predicted value in first dataset with each predicted value in second dataset.

    Total Number of pairs to compare = x * y
    x : Number of observations in first dataset (actual values of 1 in dependent variable)
    y : Number of observations in second dataset (actual values of 0 in dependent variable).

    In this step, we are performing cartesian product (cross join) of events and non-events. For example, you have 100 events and 1000 non-events. It would create 100k (100*1000) pairs for comparison.

  4. A pair is concordant if 1 (observation with the desired outcome i.e. event) has a higher predicted probability than 0 (observation without the outcome i.e. non-event).
  5. A pair is discordant if 0 (observation without the desired outcome i.e. non-event) has a higher predicted probability than 1 (observation with the outcome i.e. event).
  6. A pair is tied if 1 (observation with the desired outcome i.e. event) has same predicted probability than 0 (observation without the outcome i.e. non-event).
  7. The final percent values are calculated using the formula below -
    Percent Concordant = 100*[(Number of concordant pairs)/Total number of pairs]
    Percent Discordant = 100*[(Number of discordant pairs)/Total number of pairs]
    Percent Tied = 100*[(Number of tied pairs)/Total number of pairs]
  8. Area under curve (AUC) = (Percent Concordant + 0.5 * Percent Tied)/100
Interpretation of Concordant, Discordant and Tied Percent

Percent Concordant : Percentage of pairs where the observation with the desired outcome (event) has a higher predicted probability than the observation without the outcome (non-event). Higher the concordant number, better the model but needs to be validated on unseen data.

Percent Discordant : Percentage of pairs where the observation with the desired outcome (event) has a lower predicted probability than the observation without the outcome (non-event). Lower the discordant number, better the model but needs to be validated on unseen data.

Percent Tied : Percentage of pairs where the observation with the desired outcome (event) has same predicted probability than the observation without the outcome (non-event).

AUC : AUC is also known as c-statistics. It is calculated by adding Concordance Percent and 0.5 times of Tied Percent.

Gini coefficient or Somers' D statistic is closely related to AUC. It is calculated by (2*AUC - 1). It can also be calculated by (Percent Concordant - Percent Discordant)

In this section, you will learn how to calculate AUC using Concordance and Tied Percent in Python, SAS, and R.

Python Code

In the following code, we will use the pandas library along with the statsmodels library. You need to make sure that both libraries are installed in your Python environment. If not installed, use this command : pip install pandas pip install statsmodels

import pandas as pd
import statsmodels.formula.api as smf
import statsmodels.api as sm

# Read Data
df = pd.read_csv("https://stats.idre.ucla.edu/stat/data/binary.csv")

# Convert admit column to binary variable
df['admit'] = df['admit'].astype('int')

# Factor Variables
df['rank'] = df['rank'].astype('category')

# Logistic Model
df['rank'] = df['rank'].cat.reorder_categories([4, 1, 2, 3])
mylogistic = smf.glm(formula='admit ~ gre + gpa + rank', data=df, family=sm.families.Binomial()).fit()
print(mylogistic.summary())

# Predict
pred = mylogistic.predict()
finaldata = pd.concat([df, pd.Series(pred, name='pred')], axis=1)

def AUC(actuals, predictedScores):
    fitted = pd.DataFrame({'Actuals': actuals, 'PredictedScores': predictedScores})
    ones = fitted[fitted['Actuals'] == 1] # Subset ones
    zeros = fitted[fitted['Actuals'] == 0] # Subset zeros
    totalPairs = len(ones) * len(zeros) # calculate total number of pairs to check
    conc = sum(ones['PredictedScores'].apply(lambda x: (x > zeros['PredictedScores']).sum()))
    disc = sum(ones['PredictedScores'].apply(lambda x: (x < zeros['PredictedScores']).sum()))
    concordance = conc / totalPairs
    discordance = disc / totalPairs
    tiesPercent = (1 - concordance - discordance)
    AUC = concordance + 0.5 * tiesPercent
    Gini = 2 * AUC - 1
    return {"Concordance": concordance, "Discordance": discordance,
            "Tied": tiesPercent, "AUC": AUC, "Gini or Somers D": Gini}

print(AUC(finaldata['admit'], finaldata['pred']))

SAS Code

FILENAME PROBLY TEMP;
PROC HTTP
 URL="https://stats.idre.ucla.edu/stat/data/binary.csv"
 METHOD="GET"
 OUT=PROBLY;
RUN;

OPTIONS VALIDVARNAME=ANY;
PROC IMPORT
  FILE=PROBLY
  OUT=WORK.binary REPLACE
  DBMS=CSV;
RUN;

ods graphics on;
Proc logistic data= WORK.binary descending plots(only)=roc;
class rank / param=ref ;
model admit = gre gpa rank;
output out = estprob p= pred;
run;

/*split the data into two datasets- event and non-event*/ 
Data event nonevent;
Set estprob;
If admit = 1 then output event;
else if admit = 0 then output nonevent;
run;

/*Cartesian product of event and non-event actual cases*/ 
Proc SQL noprint;
create table pairs as
select a.admit as admit1, b.admit as admit0,
a.pred as pred1,b.pred as pred0
from event a cross join nonevent b;
quit;

/*Calculating concordant,discordant and tied percent*/
Data pairs;
set pairs;
concordant =0;
discordant=0;
tied=0;
If pred1 > pred0 then concordant = 1;
else If pred1 < pred0 then discordant = 1;
else tied = 1;
run;

/*Mean values - Final Result*/
proc sql;
select mean(Concordant)*100 as Percent_Concordant,
mean(Discordant) *100 as Percent_Discordant,
mean(Tied)*100 as Percent_Tied,
(calculated Percent_Concordant + 0.5* calculated Percent_Tied)/100 as AUC,
2*calculated AUC - 1 as somers_d
from pairs;
quit;

R Code

# Read Data
df = read.csv("https://stats.idre.ucla.edu/stat/data/binary.csv")

# Factor Variables
df$admit = as.factor(df$admit)
df$rank = as.factor(df$rank)

# Logistic Model
df$rank <- relevel(df$rank, ref='4')
mylogistic <- glm(admit ~ ., data = df, family = "binomial")
summary(mylogistic)$coefficient

# Predict
pred = predict(mylogistic, type = "response")
finaldata = cbind(df, pred)


AUC <- function (actuals, predictedScores){
  fitted <- data.frame (Actuals=actuals, PredictedScores=predictedScores)
  colnames(fitted) <- c('Actuals','PredictedScores')
  ones <- fitted[fitted$Actuals==1, ] # Subset ones
  zeros <- fitted[fitted$Actuals==0, ] # Subsetzeros
  totalPairs <- nrow (ones) * nrow (zeros) # calculate total number of pairs to check
  conc <- sum (c(vapply(ones$PredictedScores, function(x) {((x > zeros$PredictedScores))}, FUN.VALUE=logical(nrow(zeros)))), na.rm=T)
  disc <- sum(c(vapply(ones$PredictedScores, function(x) {((x < zeros$PredictedScores))}, FUN.VALUE = logical(nrow(zeros)))), na.rm = T)
  concordance <- conc/totalPairs
  discordance <- disc/totalPairs
  tiesPercent <- (1-concordance-discordance)
  AUC = concordance + 0.5*tiesPercent
  Gini = 2*AUC - 1
  return(list("Concordance"=concordance, "Discordance"=discordance,
              "Tied"=tiesPercent, "AUC"=AUC, "Gini or Somers D"=Gini))
}

AUC(finaldata$admit, finaldata$pred)
AUC and Concordance

Calculating AUC using Integration Method

Trapezoidal Rule Numerical Integration method is used to find area under curve. The area of a trapezoid is as follows:

( xi+1 – xi ) * ( yi + yi+1 ) / 2

In our case, x refers to values of false positive rate (1-Specificity) at different probability cut-offs, y refers to true positive rate (Sensitivity) at different cut-offs. Vector x needs to be sorted. Any observation with predicted probability that exceeds or equals probability cut-off is predicted to be an event; otherwise, it is predicted to be a nonevent.

( fpri+1 – fpri ) * ( tpri + tpri+1 ) / 2

fpr represents false positive rate (1- specificity). tpr represents true positive rate (sensitivity). See the image below showing step by step calculation. It includes a very few cut-offs for demonstration purpose.

AUC Calculation Steps using Integration Method

In this section, you will learn how to calculate AUC using Integration Method in Python, SAS, and R.

Python Code

import pandas as pd
import numpy as np
from sklearn.metrics import roc_curve
import statsmodels.formula.api as smf
import statsmodels.api as sm

# Read Data
df = pd.read_csv("https://stats.idre.ucla.edu/stat/data/binary.csv")

# Convert admit column to binary variable
df['admit'] = df['admit'].astype('int')

# Factor Variables
df['rank'] = df['rank'].astype('category')

# Logistic Model
df['rank'] = df['rank'].cat.reorder_categories([4, 1, 2, 3])
mylogistic = smf.glm(formula='admit ~ gre + gpa + rank', data=df, family=sm.families.Binomial()).fit()
print(mylogistic.summary())

# Predict
pred = mylogistic.predict()
finaldata = pd.concat([df, pd.Series(pred, name='pred')], axis=1)

# Calculate ROC curve
fpr, tpr, thresholds = roc_curve(finaldata['admit'], finaldata['pred'])

# Use trapezoidal rule to approximate area under ROC curve
dx = np.diff(fpr)
auroc = np.sum(dx * (tpr[1:] + tpr[:-1])) / 2
print(f'AUROC: {auroc}')

SAS Code

In the SAS program below, we are using PROC IML procedure to perform integration calculations.

FILENAME PROBLY TEMP;
PROC HTTP
 URL="https://stats.idre.ucla.edu/stat/data/binary.csv"
 METHOD="GET"
 OUT=PROBLY;
RUN;

OPTIONS VALIDVARNAME=ANY;
PROC IMPORT
  FILE=PROBLY
  OUT=WORK.binary REPLACE
  DBMS=CSV;
RUN;

ods graphics on;
Proc logistic data= WORK.binary descending plots(only)=roc;
class rank / param=ref ;
model admit = gre gpa rank / outroc=performance;
output out = estprob p= pred;
run;

proc sort data=performance;
by _1MSPEC_;
run;

proc iml;
use performance;
read all var {_SENSIT_} into sensitivity;
read all var {_1MSPEC_} into falseposrate;
N  = 2 : nrow(falseposrate);
fpr = falseposrate[N] - falseposrate[N-1];
tpr = sensitivity[N] + sensitivity[N-1];
ROC = fpr`*tpr/2;
Gini= 2*ROC - 1;
print ROC Gini;

R Code

# Read Data
df = read.csv("https://stats.idre.ucla.edu/stat/data/binary.csv")

# Factor Variables
df$admit = as.factor(df$admit)
df$rank = as.factor(df$rank)

# Logistic Model
df$rank <- relevel(df$rank, ref='4')
mylogistic <- glm(admit ~ ., data = df, family = "binomial")
summary(mylogistic)$coefficient

# Predict
pred = predict(mylogistic, type = "response")
finaldata = cbind(df, pred)

library(ROCR)
predobj <- prediction(finaldata$pred, finaldata$admit)
perf <- performance(predobj,"tpr","fpr")
plot(perf)

# Trapezoidal rule of integration
# Computes the integral of Sensitivity (Y) with respect to FalsePosRate (x)
x = perf@x.values[[1]]
y = perf@y.values[[1]]
idx = 2:length(x)
testdf=data.frame(FalsePosRate = (x[idx] - x[idx-1]), Sensitivity = (y[idx] + y[idx-1]))
(AUROC = sum(testdf$FalsePosRate * testdf$Sensitivity)/2)

Calculating AUC using Mann–Whitney U Test

Area under curve (AUC) is directly related to Mann Whitney U test. People from analytics community also call it Wilcoxon rank-sum test.

This test assumes that the predicted probability of event and non-event are two independent continuous random variables. Area under the curve = Probability that Event produces a higher probability than Non-Event. AUC=P(Event>=Non-Event)

AUC = U1/(n1 * n2) Here U1 = R1 - (n1*(n1 + 1) / 2)

where U1 is the Mann Whitney U statistic and R1 is the sum of the ranks of predicted probability of actual event. It is calculated by ranking predicted probabilities and then selecting only those cases where dependent variable is 1 and then take sum of all these cases. n1 is the number of 1s (event) in dependent variable. n2 is the number of 0s (non-events) in dependent variable.

n1*n2 is the total number of pairs (or cross product of number of events and non-events). It is similar to what we have done in concordance method to calculate AUC.

In this section, you will learn how to calculate AUC using Mann–Whitney U Test in Python, SAS, and R.

Python Code

import pandas as pd
import numpy as np
import statsmodels.formula.api as smf
import statsmodels.api as sm

# Read Data
df = pd.read_csv("https://stats.idre.ucla.edu/stat/data/binary.csv")

# Convert admit column to binary variable
df['admit'] = df['admit'].astype('int')

# Factor Variables
df['rank'] = df['rank'].astype('category')

# Logistic Model
df['rank'] = df['rank'].cat.reorder_categories([4, 1, 2, 3])
mylogistic = smf.glm(formula='admit ~ gre + gpa + rank', data=df, family=sm.families.Binomial()).fit()
print(mylogistic.summary())

# Predict
pred = mylogistic.predict()
finaldata = pd.concat([df, pd.Series(pred, name='pred')], axis=1)

# Calculate the AUC using the Mann-Whitney U test
from scipy.stats import mannwhitneyu
def auc_mann_whitney(y, pred):
    y = np.array(y, dtype=bool)
    n1 = np.sum(y)
    n2 = np.sum(~y)
    U, _ = mannwhitneyu(pred[y], pred[~y], alternative='greater')
    return U / (n1 * n2)

# Example usage
auc = auc_mann_whitney(finaldata['admit'], finaldata['pred'])
print(auc)

R Code

# Read Data
df = read.csv("https://stats.idre.ucla.edu/stat/data/binary.csv")

# Factor Variables
df$admit = as.factor(df$admit)
df$rank = as.factor(df$rank)

# Logistic Model
df$rank <- relevel(df$rank, ref='4')
mylogistic <- glm(admit ~ ., data = df, family = "binomial")
summary(mylogistic)$coefficient

# Predict
pred = predict(mylogistic, type = "response")
finaldata = cbind(df, pred)

# AUC using Mann–Whitney test
auc_mannWhitney  <- function(y, pred){
  y  <- as.logical(y)
  n1 <- sum(y)
  n2 <- sum(!y)
  R1 <- sum(rank(pred)[y])
  U1 <- R1 - n1 * (n1 + 1)/2
  U1/(n1 * n2)
}

auc_mannWhitney(as.numeric(as.character(finaldata$admit)), finaldata$pred)

SAS Code

FILENAME PROBLY TEMP;
PROC HTTP
 URL="https://stats.idre.ucla.edu/stat/data/binary.csv"
 METHOD="GET"
 OUT=PROBLY;
RUN;

OPTIONS VALIDVARNAME=ANY;
PROC IMPORT
  FILE=PROBLY
  OUT=WORK.binary REPLACE
  DBMS=CSV;
RUN;

ods graphics on;
Proc logistic data= WORK.binary descending plots(only)=roc;
class rank / param=ref ;
model admit = gre gpa rank;
output out = estprob p= pred;
run;

%let dependent_var = admit;
%let score_dataset = estprob;
%let score_column = pred;

ods output WilcoxonScores=WilcoxonScore;
proc npar1way wilcoxon data= &score_dataset.;
class &dependent_var.;
var  &score_column.;
run;

data AUC;
set WilcoxonScore end=eof;
retain v1 v2 1;
if _n_=1 then v1=abs(ExpectedSum - SumOfScores);
v2=N*v2;
if eof then do;
d=v1/v2;
Gini=d * 2;
AUC=d+0.5;   
put AUC=  GINI=;
keep AUC Gini;
output;
end;
proc print noobs;
run;

Calculating AUC using Cumulative Events and Non-Events

In this method, we will see how we can calculate area under curve using decile (binned) data.

  1. Sort predicted probabilities in descending order. It means customer having high likelihood to buy a product should appear at top (in case of propensity model)
  2. Split or rank into 10 parts. It is similar to concept of calculating decile.
  3. Calculate number of cases in each decile level. It would be same in each level as we divided the data in 10 equal parts.
  4. Calculate number of 1s (event) in each decile level. Maximum 1s should be captured in first decile (if your model is performing fine!)
  5. Calculate cumulative percent of 1s in each decile level. Last decile should have 100% as it is cumulative in nature.
  6. Similar to the above step, we will calculate cumulative percent of 0s in each decile level.
  7. AUC would be calculated using trapezoidal rule numeric integration formula. In this case, xis cumulative % of 0s and yis cumulative % of 1s
Calculate AUC using Cumulative Events and Non-Events
This method returns an approximation of AUC score since we are using only 10 bins instead of raw values.
Related Posts
Spread the Word!
Share
About Author:
Deepanshu Bhalla

Deepanshu founded ListenData with a simple objective - Make analytics easy to understand and follow. He has over 10 years of experience in data science. During his tenure, he worked with global clients in various domains like Banking, Insurance, Private Equity, Telecom and HR.

23 Responses to "A Complete Guide to Area Under Curve (AUC)"
  1. Very precise and clear explanation of concordance and discordance. Also the code helps in better understanding of the phenomenon. Thanks.

    ReplyDelete
  2. Neat explanations, really helpful to understood these definitions. Thanks!

    ReplyDelete
  3. Very clear explanation, thank you :)

    ReplyDelete
  4. Thanks for the post! Shouldn't it be proc logistic with descending option? as we are treating 1s as events and 0 as nonevents

    ReplyDelete
  5. First time I understood concordance and discordance. Thanks

    ReplyDelete
  6. For a good model what should be the concordance?

    ReplyDelete
  7. Very informative, clear, and to the point

    ReplyDelete
  8. Very good explanation and informative. Thanks Buddy keep sharing

    ReplyDelete
  9. Can you please give the calculation of concordance and disconcordance in excel format with example which will be easy to understand the calculation.

    ReplyDelete
  10. The above codes are very useful. Any suggestions for weighted data?

    ReplyDelete
  11. Hello, I want to know, what to do in cases where tied percentage is high, say 20%. How to reduce tied percentage?

    ReplyDelete
  12. Excellent Work. Thanks for such detailed description.

    ReplyDelete
  13. Thorough and very useful. However can you let me know how to derive the equation: AUC = (Percent Concordant + 0.5 * Percent Tied)/100. Basically I want to know the steps to get the above equation

    ReplyDelete
  14. Thanks for the article, but cross join is quite heavy and won't be possible on large datasets.

    ReplyDelete
  15. Really great explanation. I had one more doubt You said Gini is a special case of SomerS' D. Can you explain how are gini and Somers' D related and what is the difference ?

    ReplyDelete
  16. Your website is a god sent for students like me. Thank you for providing such easy and clear explainations.

    ReplyDelete
  17. Very few know about Percent Tied and its role in AUC. I am very thankful that you introduce it here. Could you please provide reference to the 4 methods you introduced and the proves that they are equivalent?

    ReplyDelete
Next → ← Prev