Life expectancy is a comprehensive measure of the average age at which individuals in a given population are expected to die. It encompasses a broader range of demographic factors than infant and child mortality, mainly focusing on early deaths. In pre-modern, impoverished societies, life expectancy was estimated to be around 30 years across all regions. To analyze the relevant dataset, RStudio software was employed, which is freely available and open-assessed. The data was preprocessed to ensure the absence of missing values and multicollinearity before proceeding with further analysis. Regression analysis was conducted to identify the variables that significantly impacted life expectancy. In Africa, Algeria had the highest life expectancy score at 76.88, while the Central African Republic had the lowest at 53.28. In Asia, Hong Kong recorded the highest score at 85.07, while Afghanistan had the lowest at 64.83. In Australia, Australia had the highest score at 82.90, while Papua New Guinea had the lowest at 64.50. The mean life expectancy score was computed and then replaced with mean values. Access to electricity and current health expenditure per capita impacted life expectancy significantly. The mortality rate harmed life expectancy, with higher mortality rates leading to lower life expectancy scores and vice versa. Unemployment and newly infected HIV also had an impact on life expectancy.

Descriptive Statistics

The descriptive statistics results showed that some all column contain missing value that ranged from 1 to 195. The mean value was for all variable was EG.ELC.ACCS.ZS was 86.47, EN.POP.DNST (446.04), FR.INR.RINR (6.22), NY.ADJ.NNTY.KD.ZG (4.03), NY.ADJ.NNTY.PC.KD.ZG (2.66), NY.GDP.MKTP.KD.ZG (2.81), NY.GDP.PCAP.CD (18605.51), SE.ADT.LITR.ZS (92.04), SE.COM.DURS (9.92), SE.PRM.CMPT.ZS (93.05), SE.PRM.CUAT.ZS (87.74), SE.PRM.UNER (98649.65), SE.TER.CUAT.BA.ZS (19.86), SH.H2O.SMDW.ZS (73.70), SH.HIV.INCD (7574.42), SH.HIV.INCD.14 (1650.44), SH.XPD.CHEX.GD.ZS (6.60), SH.XPD.CHEX.PC.CD (1143.71), SI.POV.LMIC (10.13), SL.UEM.TOTL.NE.ZS (7.67), SP.DYN.CBRT.IN (19.37), SP.DYN.IMRT.IN (20.97), SP.DYN.LE00.IN (72.93), SP.POP.GROW(1.19) and SP.POP.TOTL (35445759.77)

Data visualization

Comparison of Life expectancy in continent

In our data set there was six continent in our data. The highest life expectancy was observed in Europe that was 79.3, followed by North and South America was 76.2, and 75.1 respectively, Asia was 74.6, Australia was 73.5 and Africa that was 64.1. The highest life expectancy was observed in Europe

Comparison of Life expectancy in Country

In the second graph, life expectancy was found highest in chile that was 80.2, followed by Uruguay was 77.9, Colombia was 77.3. The lowest value of expectancy was 69.9 in Guyana in South America continent

Comparison of Life expectancy in country with continent

Regression model

The full regression model was run along with life expectancy and all other variables as an independent variable while country name, code and continent variables was excluded in the data set. EG.FEC.RNEW.ZS was removed in the full model due to the singularity. The regression coefficient analysis showed that intercept value was found significant along with EG.ELC.ACCS. ZS the coefficient value was found significant and positive that means this variable leads to increased in the value of life expectancy value. SP.POP.GROW and SH.XPD.CHEX.PC.CD, SL.UEM.TOTL.NE.ZS was also found positive and significant. SH.HIV.INCD and SP.DYN.IMRT.IN found negative and significant that means these variables leads to decreased in the life expectancy value. The adjusted R square value was 0.87 that means our model fitness was good as shown in Supplementary S1. Through VIF multicollinearity problem was detected. The variance inflation factor (VIF), which evaluates the correlation and intensity of correlation between the predictor variables in a regression model, is the most frequent approach to discover multicollinearity. The results showed that these two variables that are NY.ADJ.NNTY.KD.ZG and NY.ADJ.NNTY.PC.KD.ZG have highest multicollinearity issue that means these variables was significantly correlated.

All non-significant variables was removed from the model and also variables was removed which have multicollinearity issue. In the final model continent variable was added in the model. The coefficient of the model showed that Asia continent have significant impact on the life expectancy with positive values. Life expectancy was significantly affected by the access to electricity and also found significant, from the results it was concluded that people who have access to electricity have high life expectancy. Mortality rate had negative impact on the life expectancy score, when the mortality rate was high life expectancy was decreased and vice versa. Current health expenditure per capita, PPP (current international \$) also have significant positive impact on the life expectancy score and found significant. Unemployment and newly infected HIV have negative impact on the life expectancy score. The model fitness was also found high that was 0.864

Conclusion

In Africa Algeria have highest value that was 76.88 and Central African Republic that was 53.28. In Asia Hing Kong the value was 85.07 and Afghanistan that was 64.83. In Australia Continent, Australia that was 82.90 and Papua New Guinea that is 64.50. There is a lot of missing value in the data set as shown in table 1. If we remove the missing value, then all observation was removed and that was not a good option. So, the mean value of life expectancy was computed and then replaced with mean values. Life expectancy was significantly affected by the access to electricity and current health expenditure per capita. Mortality rate had negative impact on the life expectancy score, when the mortality rate was high life expectancy was decreased and vice versa. Unemployment and newly infected HIV have an impact on life expectancy.

Another Data Visualiztion Project