"from all angles in order to to explain what's wrong."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Comments and further Analysis"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"While there this analysis is well structured, there are few choices that may have affected the conclusions:\n",
"\n",
" 1. Removing the samples where there was no malfunction was not well-justified and may have introduced bias in the results.\n",
" 2. The analysis did not consider the effect of pressure on the outcomes.\n",
" 3. The sample size is very small making it difficult to completely rely on the results of the analysis. Given the magnitude of this project and the fact that human life was at stake, the engineers should have been more cautious.\n",
"\n",
"In the following sections, we will perform further analysis based on the comment 1 above."
"With the full data set, the most likely estimator of the temparature is $-0.1156$ indicating a negative correlation between the temparature and the risk of failure. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Estimation of the probability of O-ring malfunction"
"The new plot shows that the probability of failure at $31^oF$ is more that $80\\%$. This is in stark constrast to the results of our previous analysis that put the probability at $20\\%$. This analysis shows that indeed there is a negative correlation between temparature and possibility and failure and it is not correct to exclude data from successful launches."
