RPubs link information

• The RPub link is given below: http://rpubs.com/s3775152_/tamil96

Introduction

• Over a period of time, the number of people affected by diseases has gradually increased.
• Specifically number of people suffering from liver disease has increased because of excessive consumption of alcohol, inhale of harmful gases, intake of contaminated food, pickles and drugs.
• These habits have been noticed across people of different age groups and gender, hence we want to know to what extend does age and gender effect the chances of have liver disease.

Introduction Cont.

• A dataset having different parameters related to Indian patients suffering from different disease is considered.
• To conduct analysis, a hypothesis test is going to be conducted considering age and gender.
• We expect to gain conclusions based on the mean age difference in the genders (Male and Female) of Indian liver patients.

Problem Statement

• In order to conduct an investigation on the difference in the mean age of the Indian patients suffering from liver disease which can give results to determine which of the two genders Male or Female will suffer at the earliest from the fatal disease of the liver.

-To test if there is any difference in the mean age in case of male and female we will be using statistical hypothesis.

• In the above considered case the null hypothesis is that there is no significant mean age difference in case of Indian male and female patients of liver disease.
  Ho : μ1 = μ2

• The alternate hypothesis is that there is significant mean age difference in case of Indian male and female patients of liver disease. HA : μ1 ≠ μ2

• Due to presence of the group Male and Female we have later used two sample test to test this hypothesis. This will conclude if the population is independent of each other.

Data

The dataset selected for this investigation is from Kaggle.com. This dataset is an open dataset and has data for indian liver patient records. It has patient records collected from North East of Andhra Pradesh, India.

• Link for dataset: https://www.kaggle.com/uciml/indian-liver-patient-records

• Number of observations: 583
• Number of attributes: 11

• Number of missing values: 0

Data Cont.

• The variables available in the dataset are Age, Gender, Total_BilirubinTotal, Direct_Bilirubin, Alkaline_Phosphotase, Alamine_Aminotransferase, Aspartate_Aminotransferase, Total_Protiens, Albumin, Albumin_and_Globulin_Ratio, Dataset

• The attributes from the dataset used for analysis is Age and Gender
• The factor variables are male and female.

• In the dataset the range of age is 4years to 90years

Descriptive Statistics and Visualisation

• In the dataset to carry further analysis the gender which was defined as character initially was changed to factor to simplify the analysis.

• In the process of further analysis we intend to determine the number of NA values and there was found to be none of them.

• Summarise the important variables in your investigation.
• Use visualisation to highlight interesting features of the data and tell the overall story.
• Explain how you dealt with data issues (if any), e.g. missing data and outliers.

• Here are the examples of R chunks and outputs

library(readr)
indian_liver_patient<-read_csv("indian_liver_patient.csv")
View(indian_liver_patient)

indian_liver_patient$Gender = factor(indian_liver_patient$Gender , ordered = TRUE )
str(indian_liver_patient)

## Classes 'spec_tbl_df', 'tbl_df', 'tbl' and 'data.frame': 583 obs. of  11 variables:
##  $ Age                       : num  65 62 62 58 72 46 26 29 17 55 ...
##  $ Gender                    : Ord.factor w/ 2 levels "Female"<"Male": 1 2 2 2 2 2 1 1 2 2 ...
##  $ Total_Bilirubin           : num  0.7 10.9 7.3 1 3.9 1.8 0.9 0.9 0.9 0.7 ...
##  $ Direct_Bilirubin          : num  0.1 5.5 4.1 0.4 2 0.7 0.2 0.3 0.3 0.2 ...
##  $ Alkaline_Phosphotase      : num  187 699 490 182 195 208 154 202 202 290 ...
##  $ Alamine_Aminotransferase  : num  16 64 60 14 27 19 16 14 22 53 ...
##  $ Aspartate_Aminotransferase: num  18 100 68 20 59 14 12 11 19 58 ...
##  $ Total_Protiens            : num  6.8 7.5 7 6.8 7.3 7.6 7 6.7 7.4 6.8 ...
##  $ Albumin                   : num  3.3 3.2 3.3 3.4 2.4 4.4 3.5 3.6 4.1 3.4 ...
##  $ Albumin_and_Globulin_Ratio: num  0.9 0.74 0.89 1 0.4 1.3 1 1.1 1.2 1 ...
##  $ Dataset                   : num  1 1 1 1 1 1 1 1 2 1 ...
##  - attr(*, "spec")=
##   .. cols(
##   ..   Age = col_double(),
##   ..   Gender = col_character(),
##   ..   Total_Bilirubin = col_double(),
##   ..   Direct_Bilirubin = col_double(),
##   ..   Alkaline_Phosphotase = col_double(),
##   ..   Alamine_Aminotransferase = col_double(),
##   ..   Aspartate_Aminotransferase = col_double(),
##   ..   Total_Protiens = col_double(),
##   ..   Albumin = col_double(),
##   ..   Albumin_and_Globulin_Ratio = col_double(),
##   ..   Dataset = col_double()
##   .. )

colSums(is.na(indian_liver_patient))

##                        Age                     Gender 
##                          0                          0 
##            Total_Bilirubin           Direct_Bilirubin 
##                          0                          0 
##       Alkaline_Phosphotase   Alamine_Aminotransferase 
##                          0                          0 
## Aspartate_Aminotransferase             Total_Protiens 
##                          0                          0 
##                    Albumin Albumin_and_Globulin_Ratio 
##                          0                          4 
##                    Dataset 
##                          0

Decsriptive Statistics Cont.

• For the next step we filter out male and female factor variables.

• Next we use the summarize function to find out the quartiles, median, mean, maximum, minimum and standard deviation for each gender separately.

• In order to figure out if there is any difference in the mean age visually we use box plot

• You can use the knitr:kable function to print nice HTML tables. Here is an example R code:

indian_liver_patient_Male = indian_liver_patient %>% filter(Gender == 'Male')

indian_liver_patient_Male %>% summarise(       Min = min(indian_liver_patient_Male$Age, na.rm = TRUE),       Q1 = quantile(indian_liver_patient_Male$Age, probs = .25, na.rm = TRUE),       Median = median(indian_liver_patient_Male$Age, na.rm = TRUE),       Q3 = quantile(indian_liver_patient_Male$Age, probs = .75, na.rm = TRUE),       Max = max(indian_liver_patient_Male$Age, na.rm = TRUE),       Mean = mean(indian_liver_patient_Male$Age, na.rm = TRUE),       SD = sd(indian_liver_patient_Male$Age, na.rm = TRUE),       n = n(),       Missing = sum(is.na(indian_liver_patient_Male$Age))     ) 

indian_liver_patient_Female = indian_liver_patient %>% filter(Gender =='Female')

indian_liver_patient_Female %>% summarise(       Min = min(indian_liver_patient_Female$Age, na.rm = TRUE),       Q1 = quantile(indian_liver_patient_Female$Age, probs = .25, na.rm = TRUE),       Median = median(indian_liver_patient_Female$Age, na.rm = TRUE),       Q3 = quantile(indian_liver_patient_Female$Age, probs = .75, na.rm = TRUE),       Max = max(indian_liver_patient_Female$Age, na.rm = TRUE),       Mean = mean(indian_liver_patient_Female$Age, na.rm = TRUE),       SD = sd(indian_liver_patient_Female$Age, na.rm = TRUE),       n = n(),       Missing = sum(is.na(indian_liver_patient_Female$Age))     ) 

indian_liver_patient %>% boxplot(Age ~ Gender,data = ., ylab = "AGE", main = "BOX PLOT OF MALE AND FEMALE" , ylim= c(5,100), col = "pink")

Hypothesis Testing

• In order to verify the assumption of equal variances i.e. to assure the homogeneity of variances across parameters Age and Gender we use the levene’s test.
• Next we use the two sample t-test assuming equal variance based on the outcome of levene’s test.

library(car)
leveneTest(Age ~ Gender , data= indian_liver_patient)

t.test(Age ~ Gender , data = indian_liver_patient , var.equal =  TRUE, alternative = "two.sided")

## 
##  Two Sample t-test
## 
## data:  Age by Gender
## t = -1.3655, df = 581, p-value = 0.1726
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.197309  0.934302
## sample estimates:
## mean in group Female   mean in group Male 
##             43.13380             45.26531

Hypthesis Testing Cont.

• Here are the examples of mathematical equations:

\[H_0: \mu_1 = \mu_2 \]

\[H_A: \mu_1 \ne \mu_2\]

\[S = \sum^n_{i = 1}d^2_i\]

• You can also place equations inline: \(z = \frac{x - \mu}{\sigma}\)

Discussion

• In the levene’s test the p-value obtained is 0.243 which is greater than 0.05, hence we fail to reject Ho and assume equal variances.

-Further in the two sample t-test assumimg equal variances the p-value is 0.1726 which is greater than 0.05 hence we fail to reject the null hypothesis and accept Ho, that there is no significant mean age difference in case of Indian male and female patients of liver disease.

• The mean age of Male is 45.26531 and mean age of female is 43.1338.

• After conducting appropriate analysis it can be concluded that Male with age 45.26531 and Female with age 43.1338 are likely to suffer from liver disease.

• The findings of this investigation needs to be an eye-opener to people all around. Males and Females exposed to the suffering of lung disease at such early age is likely to be a matter of concern to the society. Also this will affect the next generation as they are likely to be influenced by the habits followed by elders. Proper initiatives need to be taken to spread awareness along with precautionary measures regarding this issue.

References

• Indian Liver Patient Records Patient records collected from North East of Andhra Pradesh, India https://www.kaggle.com/uciml/indian-liver-patient-records