publicdata/j40-cejst-2
1
0
Fork 0
mirror of https://github.com/DOI-DO/j40-cejst-2.git synced 2025-02-23 10:04:18 -08:00
Code Issues Projects Releases Packages Wiki Activity Actions

comments

Browse source
This commit is contained in:
Saran Ahluwalia 2021-12-11 13:40:13 -05:00
parent f7e3237ed9
commit 5ce32ccd50
1 changed files with 30 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

33
data/data-pipeline/data_pipeline/ipython/hud_eda_se_12_11_2011_relative_differences_between_methodologies.ipynb
View file

@ -496,6 +496,13 @@
"housingburden.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### As desired we see a uniform distribution for the percentile rank for burdened households"
]
},
{
"cell_type": "code",
"execution_count": 28,
@ -535,6 +542,13 @@
"sns.histplot(housingburden[\"hbrd_rank\"])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Now we compute for a baseline comparison "
]
},
{
"cell_type": "code",
"execution_count": 20,
@ -750,13 +764,20 @@
},
{
"cell_type": "code",
"execution_count": 66,
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"housingburden.to_csv(\"housing_burden.csv\", index=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Now we construct the distribution of differences in the number of owned and rented burdened households\n"
]
},
{
"cell_type": "code",
"execution_count": 27,
@ -786,7 +807,6 @@
}
],
"source": [
"### Now compare to relative \n",
"\n",
"plt.figure(figsize=(12, 8))\n",
"plt.title('Distribution of differences between two methodologies')\n",
@ -802,7 +822,7 @@
"cell_type": "markdown",
"metadata": {},
"source": [
"### Below are KDE plots of the respective densities. Green is the current methodology; red represents the revision."
"### Plotting both distributions for accounting for different aggregations of owned and rented burdened households. Red is the revised version; green is the current methodology"
]
},
{
@ -1040,6 +1060,13 @@
"p = np.sum(bs_replicates >= empirical_diff_means) / len(bs_replicates)\n",
"print(\"P Value: {:.4f}\".format(round(p, 2)))"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {