From f943a2fe9d3a60fbde44c9f6b8e453f7dda7390a Mon Sep 17 00:00:00 2001
From: NourElh <734092651fcdd5add927271f472626a6@app-learninglab.inria.fr>
Date: Wed, 2 Nov 2022 12:45:23 +0000
Subject: [PATCH] Update exo5_en.Rmd

---
 module2/exo5/exo5_en.Rmd | 26 ++++++++++++++++++++++++++
 1 file changed, 26 insertions(+)

diff --git a/module2/exo5/exo5_en.Rmd b/module2/exo5/exo5_en.Rmd
index f9003e3..7dc084b 100644
--- a/module2/exo5/exo5_en.Rmd
+++ b/module2/exo5/exo5_en.Rmd
@@ -30,14 +30,27 @@ data = read.csv("shuttle.csv",header=T)
 data
 ```
 
+
 The data set shows us the date of each test, the number of O-rings
 (there are 6 on the main launcher), the
 temperature (in Fahrenheit) and pressure (in psi), and finally the
 number of identified malfunctions.
 
+### Comment: 
+There is no datapoint for low temperatures as the one used during launching =>  Lack of data.
+
 # Graphical inspection
 Flights without incidents do not provide any information
 on the influence of temperature or pressure on malfunction.
+### New analysis:
+However, I'd like to visualize the data of successful flights.
+```{r}
+d = data[data$Malfunction==0,]
+d
+```
+We can see that for high temperatures, there are failures, so excluding this data may prevent the detection of the origin of the failures.
+
+### Continue
 We thus focus on the experiments in which at least one O-ring was defective.
 
 ```{r}
@@ -78,6 +91,14 @@ and the standard error of this estimator is 0.049, in other words we
 cannot distinguish any particular impact and we must take our
 estimates with caution.
 
+### New analysis
+```{r}
+library(ggplot2)
+ggplot(d,aes(x=Temperature, y=Malfunction/Count))+geom_point(alpha=.3,size=3)+theme_bw()+geom_smooth(method = "glm",method.args=list(family="binomial"))+xlim(0,200)
+```
+
+From the graph, we see that there is a big uncertainty of this estimate, so the possibility that temperature has an effect cannot be ruled out.
+
 # Estimation of the probability of O-ring malfunction
 The expected temperature on the take-off day is 31°F. Let's try to
 estimate the probability of O-ring malfunction at
@@ -85,6 +106,8 @@ this temperature from the model we just built:
 
 ```{r}
 # shuttle=shuttle[shuttle$r!=0,] 
+logistic_reg = glm(data=data, Malfunction/Count ~ Temperature, weights=Count, 
+                   family=binomial(link='logit'))
 tempv = seq(from=30, to=90, by = .5)
 rmv <- predict(logistic_reg,list(Temperature=tempv),type="response")
 plot(tempv,rmv,type="l",ylim=c(0,1))
@@ -117,3 +140,6 @@ of NASA, which have a lot to do with this fiasco, the previous analysis
 includes (at least) a small problem.... Can you find it?
 You are free to modify this analysis and to look at this dataset
 from all angles in order to to explain what's wrong.
+
+### Comment:
+Lack of low temperatures data + aggregating data + not considering the uncertainty of the estimate have led to this mistake.
-- 
2.18.1