"
],
"text/plain": [
" State State_Code County County_Code Year \\\n",
"29268 South Carolina SC SPARTANBURG 45083 2013 \n",
"5670 Georgia GA JONES 13169 2015 \n",
"13463 Kentucky KY MARION 21155 2008 \n",
"6463 Georgia GA UPSON 13293 2015 \n",
"31212 Texas TX CAMERON 48061 2003 \n",
"\n",
" State_Mortality_Rate Population_x Deaths Mortality_Rate _merge \\\n",
"29268 0.000125 290545.0 35 0.00012 both \n",
"5670 0.000122 NaN left_only \n",
"13463 0.000193 NaN left_only \n",
"6463 0.000122 NaN left_only \n",
"31212 0.000068 NaN left_only \n",
"\n",
" Original Population_y merge2 \n",
"29268 True 290545 both \n",
"5670 False 28432 both \n",
"13463 False 19647 both \n",
"6463 False 26237 both \n",
"31212 False 358492 both "
]
},
"execution_count": 35,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Remapping with population data\n",
"df6 = pd.merge(\n",
" df5,\n",
" population[[\"County_Code\", \"Year\", \"Population\"]],\n",
" on=[\"County_Code\", \"Year\"],\n",
" how=\"left\",\n",
" indicator=\"merge2\",\n",
" validate=\"1:1\",\n",
")\n",
"df6.sample(5)"
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"merge2\n",
"both 38673\n",
"left_only 0\n",
"right_only 0\n",
"Name: count, dtype: int64"
]
},
"execution_count": 36,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df6[\"merge2\"].value_counts()"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"count 3.867300e+04\n",
"mean 1.016857e+05\n",
"std 3.201735e+05\n",
"min 0.000000e+00\n",
"25% 1.229900e+04\n",
"50% 2.715300e+04\n",
"75% 6.946000e+04\n",
"max 1.007726e+07\n",
"Name: Population_y, dtype: float64"
]
},
"execution_count": 37,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# check if we have population data for all counties\n",
"df6[\"Population_y\"].describe()"
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# graph to check missing in original vs population\n",
"threshold = []\n",
"missing = []\n",
"for i in range(0, 300000, 5000):\n",
" threshold.append(i)\n",
" missing.append(\n",
" (df6[df6[\"Population_y\"] >= i][\"Mortality_Rate\"].isna().sum()) / len(df6) * 100\n",
" )\n",
"\n",
"# plotting this\n",
"plt.plot(threshold, missing)\n",
"plt.xlabel(\"Minimum County Population Threshold\")\n",
"plt.ylabel(\"Missing Values in %\")\n",
"plt.title(\"Missing Values vs County Population Threshold\")\n",
"\n",
"# Saving the plot\n",
"plt.savefig(\"../.01_Data/Missing_vs_Population.png\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The lower population counties will add a lot of noise to the data, hence it is important to consider this when we plan to use the imputed data for any analysis."
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [],
"source": [
"# function to Impute the deaths\n",
"def new_death(row):\n",
" if pd.isna(row[\"Deaths\"]):\n",
" # calculate based on state mortality rate, with upper theshold of 9\n",
" return min(int(row[\"Population_y\"] * row[\"State_Mortality_Rate\"]), 9)\n",
" else:\n",
" # has original data so return the original value\n",
" return row[\"Deaths\"]"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
State
\n",
"
State_Code
\n",
"
County
\n",
"
County_Code
\n",
"
Year
\n",
"
State_Mortality_Rate
\n",
"
Population_x
\n",
"
Deaths
\n",
"
Mortality_Rate
\n",
"
_merge
\n",
"
Original
\n",
"
Population_y
\n",
"
merge2
\n",
"
Deaths_2
\n",
"
\n",
" \n",
" \n",
"
\n",
"
17930
\n",
"
Mississippi
\n",
"
MS
\n",
"
COPIAH
\n",
"
28029
\n",
"
2014
\n",
"
0.00015
\n",
"
NaN
\n",
"
<NA>
\n",
"
<NA>
\n",
"
left_only
\n",
"
False
\n",
"
28976
\n",
"
both
\n",
"
4
\n",
"
\n",
"
\n",
"
12320
\n",
"
Kansas
\n",
"
KS
\n",
"
STEVENS
\n",
"
20189
\n",
"
2009
\n",
"
0.000083
\n",
"
NaN
\n",
"
<NA>
\n",
"
<NA>
\n",
"
left_only
\n",
"
False
\n",
"
5592
\n",
"
both
\n",
"
0
\n",
"
\n",
"
\n",
"
3895
\n",
"
Florida
\n",
"
FL
\n",
"
DUVAL
\n",
"
12031
\n",
"
2008
\n",
"
0.000135
\n",
"
855437.0
\n",
"
89
\n",
"
0.000104
\n",
"
both
\n",
"
True
\n",
"
855437
\n",
"
both
\n",
"
89
\n",
"
\n",
"
\n",
"
14815
\n",
"
Louisiana
\n",
"
LA
\n",
"
WEST CARROLL
\n",
"
22123
\n",
"
2008
\n",
"
0.000145
\n",
"
NaN
\n",
"
<NA>
\n",
"
<NA>
\n",
"
left_only
\n",
"
False
\n",
"
11670
\n",
"
both
\n",
"
1
\n",
"
\n",
"
\n",
"
28304
\n",
"
Pennsylvania
\n",
"
PA
\n",
"
LEHIGH
\n",
"
42077
\n",
"
2014
\n",
"
0.000197
\n",
"
357591.0
\n",
"
55
\n",
"
0.000154
\n",
"
both
\n",
"
True
\n",
"
357591
\n",
"
both
\n",
"
55
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" State State_Code County County_Code Year \\\n",
"17930 Mississippi MS COPIAH 28029 2014 \n",
"12320 Kansas KS STEVENS 20189 2009 \n",
"3895 Florida FL DUVAL 12031 2008 \n",
"14815 Louisiana LA WEST CARROLL 22123 2008 \n",
"28304 Pennsylvania PA LEHIGH 42077 2014 \n",
"\n",
" State_Mortality_Rate Population_x Deaths Mortality_Rate _merge \\\n",
"17930 0.00015 NaN left_only \n",
"12320 0.000083 NaN left_only \n",
"3895 0.000135 855437.0 89 0.000104 both \n",
"14815 0.000145 NaN left_only \n",
"28304 0.000197 357591.0 55 0.000154 both \n",
"\n",
" Original Population_y merge2 Deaths_2 \n",
"17930 False 28976 both 4 \n",
"12320 False 5592 both 0 \n",
"3895 True 855437 both 89 \n",
"14815 False 11670 both 1 \n",
"28304 True 357591 both 55 "
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df7 = df6.copy()\n",
"df7[\"Deaths_2\"] = df7.apply(new_death, axis=1)\n",
"df7.sample(5)"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"