Trans Boundary Infrastructure Footprint (TBIF) Based Green House Gas Accounting

• Ankit Jain(11120018)
• Sachin (11113094)
• Shantanu Agarwal(11113099)
• Sahil Singhal(11113095)
• Rashesh Gupta(11113086)

Date: November 23, 2014

[1] "kWh.HH.mo"

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
    253     281     310     341     386     461 

[1] "L.LPG.HH.mo"

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
   18.8    30.7    42.5    47.0    61.2    79.8 

data <- read.csv(file = "tbif.csv", header=TRUE);
library(datasets); require(stats); require(graphics);
data <-data[1:3,2:18];
pairs(data,panel = panel.smooth, main = "TBIF",pch=21,bg = c("red", "green", "blue"));

y<-data$Total.Commercial.industrial.intensity.MJ.
GDP.yr..end.use.;
plot(data$Electricity.EF..kg.CO2.eq.kWh.,
data$Total.Commercial.industrial.intensity.MJ
.GDP.yr..end.use.,bg =  c("red", "green", "blue"),type="p", 
col = "black", cex = 1.1, pch = 21,frame = FALSE, xlab="Electricity", ylab = "Total.Commercial.industrial.intensity")
abline(lm(Total.Commercia.industrial.intensity.MJ
.GDP.yr..end.use.~ Electricity.EF..kg.CO2.eq.kWh., data = data), col="red")
legend("right", c("Delhi","Mumbai","Chandigarh"), pch = 21,pt.bg=c("red","green","blue"),bg="white")

cor(data$Water..treated.water.WW..1000.liters.capita.yr. 
    , data$L.LPG.HH.mo)

[1] 0.9556

fit <- lm(Surface.travel.intensity..VKT.capita.day. ~ L.kerosene.HH.mo, data = data)

summary(fit)

Call:
lm(formula = Surface.travel.intensity..VKT.capita.day. ~ L.kerosene.HH.mo, 
    data = data)

Residuals:
      1       2       3 
 0.0221 -0.0426  0.0205 

Coefficients:
                 Estimate Std. Error t value Pr(>|t|)   
(Intercept)       28.0333     0.3402    82.4   0.0077 **
L.kerosene.HH.mo  -2.0975     0.0329   -63.7   0.0100 **
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.0522 on 1 degrees of freedom
Multiple R-squared:     1,  Adjusted R-squared:     1 
F-statistic: 4.05e+03 on 1 and 1 DF,  p-value: 0.01

coef(fit)

     (Intercept) L.kerosene.HH.mo 
          28.033           -2.098 

e <- resid(fit)
yhat <- predict(fit)
max(abs(e - (y  - yhat)))

[1] 7.93

data<-read.csv(file = "multiplied_by_ef.csv", header=TRUE);
data <-  data[1:3,2:8];
fit <- lm(Total.GHGs ~ . , data = data)
fit1 <- aov(Total.GHGs~kWh.HH.mo,data=data)
fit1

Call:
   aov(formula = Total.GHGs ~ kWh.HH.mo, data = data)

Terms:
                kWh.HH.mo Residuals
Sum of Squares      61422       736
Deg. of Freedom         1         1

Residual standard error: 27.12
Estimated effects may be unbalanced

fit1 <- lm(Total.GHGs~kWh.HH.mo,data=data)
fit1

Call:
lm(formula = Total.GHGs ~ kWh.HH.mo, data = data)

Coefficients:
(Intercept)    kWh.HH.mo  
     -62.25         1.63  

fit2 <- aov(Total.GHGs~kWh.HH.mo+L.LPG.HH.mo
            ,data=data)
fit2

Call:
   aov(formula = Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo, data = data)

Terms:
                kWh.HH.mo L.LPG.HH.mo
Sum of Squares      61422         736
Deg. of Freedom         1           1

Estimated effects may be unbalanced

anova(fit,fit1,fit2,fit3,fit4,fit5,fit6)

Analysis of Variance Table

Model 1: Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo + L.kerosene.HH.mo + Industrial.process..t.waste.capita.yr. + 
    Air.Travel..L.jet.fuel.enplaned.passenger + Water..treated.water.WW..1000.liters.capita.yr.
Model 2: Total.GHGs ~ kWh.HH.mo
Model 3: Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo
Model 4: Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo + L.kerosene.HH.mo
Model 5: Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo + L.kerosene.HH.mo + Industrial.process..t.waste.capita.yr.
Model 6: Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo + L.kerosene.HH.mo + Industrial.process..t.waste.capita.yr. + 
    Air.Travel..L.jet.fuel.enplaned.passenger
Model 7: Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo + L.kerosene.HH.mo + Industrial.process..t.waste.capita.yr. + 
    Air.Travel..L.jet.fuel.enplaned.passenger + Water..treated.water.WW..1000.liters.capita.yr.
  Res.Df RSS Df Sum of Sq F Pr(>F)
1      0   0                      
2      1 736 -1      -736         
3      0   0  1       736         
4      0   0  0         0         
5      0   0  0         0         
6      0   0  0         0         
7      0   0  0         0         

summary(fit2)

Call:
lm(formula = Total.GHGs ~ kWh.HH.mo + L.LPG.HH.mo, data = data)

Residuals:
ALL 3 residuals are 0: no residual degrees of freedom!

Coefficients:
            Estimate Std. Error t value Pr(>|t|)
(Intercept)   -83.49         NA      NA       NA
kWh.HH.mo       1.60         NA      NA       NA
L.LPG.HH.mo     0.63         NA      NA       NA

Residual standard error: NaN on 0 degrees of freedom
Multiple R-squared:     1,  Adjusted R-squared:   NaN 
F-statistic:  NaN on 2 and 0 DF,  p-value: NA

Two-variable Regression - To find the amount of GHGs contribution of an unknown variable when its relationship with another parameter is known

Multi-Variate Regression - To find the overall GHG production of a city or state when various contributors to GHGs are known

• Many More variables in the model
• Using data of various cities and state to make a much better model
• Also, using better regression by going into higher order regression modelling

We would like to thank Dr. B.R. Gurjar to give us and guide us in this project.

Thanks to Ajay Nagapure sir for his constatnt guidance by quick email replies

Working on this project to make a better and much more reliable model. Also, publishing research papers along the way.