Skip to content

Instantly share code, notes, and snippets.

@GStechschulte

GStechschulte/plot_slopes_test.ipynb

Created July 16, 2023 18:18
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save GStechschulte/bb6d67cb12e00be8d69f00144b0d18e1 to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save GStechschulte/bb6d67cb12e00be8d69f00144b0d18e1 to your computer and use it in GitHub Desktop.
Download ZIP
slopes and plot_slopes core numeric variable functionality
Display the source blob
Display the rendered blob
Raw
plot_slopes_test.ipynb
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import arviz as az\n",
"import numpy as np\n",
"import pandas as pd\n",
"import seaborn.objects as so\n",
"import seaborn as sns\n",
"import matplotlib.pyplot as plt\n",
"\n",
"import bambi as bmb\n",
"from bambi.plots import plot_cap, slopes, plot_slopes, plot_comparison, comparisons\n",
"\n",
"%load_ext autoreload\n",
"%autoreload 2"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"data = bmb.load_data(\"mtcars\")\n",
"data[\"cyl\"] = data[\"cyl\"].replace({4: \"low\", 6: \"medium\", 8: \"high\"})\n",
"data[\"gear\"] = data[\"gear\"].replace({3: \"A\", 4: \"B\", 5: \"C\"})\n",
"data[\"cyl\"] = pd.Categorical(data[\"cyl\"], categories=[\"low\", \"medium\", \"high\"], ordered=True)"
]
},
{
"cell_type": "code",
"execution_count": 175,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>mpg</th>\n",
" <th>cyl</th>\n",
" <th>disp</th>\n",
" <th>hp</th>\n",
" <th>drat</th>\n",
" <th>wt</th>\n",
" <th>qsec</th>\n",
" <th>vs</th>\n",
" <th>am</th>\n",
" <th>gear</th>\n",
" <th>carb</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>21.0</td>\n",
" <td>medium</td>\n",
" <td>160.0</td>\n",
" <td>110</td>\n",
" <td>3.9</td>\n",
" <td>2.620</td>\n",
" <td>16.46</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>B</td>\n",
" <td>4</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>21.0</td>\n",
" <td>medium</td>\n",
" <td>160.0</td>\n",
" <td>110</td>\n",
" <td>3.9</td>\n",
" <td>2.875</td>\n",
" <td>17.02</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>B</td>\n",
" <td>4</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" mpg cyl disp hp drat wt qsec vs am gear carb\n",
"0 21.0 medium 160.0 110 3.9 2.620 16.46 0 1 B 4\n",
"1 21.0 medium 160.0 110 3.9 2.875 17.02 0 1 B 4"
]
},
"execution_count": 175,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data.head(2)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Auto-assigning NUTS sampler...\n",
"Initializing NUTS using jitter+adapt_diag...\n",
"Multiprocess sampling (2 chains in 4 jobs)\n",
"NUTS: [mpg_sigma, Intercept, hp, wt, hp:wt, drat]\n"
]
},
{
"data": {
"text/html": [
"\n",
"<style>\n",
" /* Turns off some styling */\n",
" progress {\n",
" /* gets rid of default border in Firefox and Opera. */\n",
" border: none;\n",
" /* Needs to be in here for Safari polyfill so background images work as expected. */\n",
" background-size: auto;\n",
" }\n",
" progress:not([value]), progress:not([value])::-webkit-progress-bar {\n",
" background: repeating-linear-gradient(45deg, #7e7e7e, #7e7e7e 10px, #5c5c5c 10px, #5c5c5c 20px);\n",
" }\n",
" .progress-bar-interrupted, .progress-bar-interrupted::-webkit-progress-bar {\n",
" background: #F44336;\n",
" }\n",
"</style>\n"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"\n",
" <div>\n",
" <progress value='4000' class='' max='4000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
" 100.00% [4000/4000 00:04&lt;00:00 Sampling 2 chains, 0 divergences]\n",
" </div>\n",
" "
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"Sampling 2 chains for 1_000 tune and 1_000 draw iterations (2_000 + 2_000 draws total) took 5 seconds.\n",
"We recommend running at least 4 chains for robust computation of convergence diagnostics\n"
]
}
],
"source": [
"model = bmb.Model(\"mpg ~ hp * wt + drat\", data=data, family=\"gaussian\")\n",
"idata = model.fit(draws=1000, chains=2)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>mean</th>\n",
" <th>sd</th>\n",
" <th>hdi_3%</th>\n",
" <th>hdi_97%</th>\n",
" <th>mcse_mean</th>\n",
" <th>mcse_sd</th>\n",
" <th>ess_bulk</th>\n",
" <th>ess_tail</th>\n",
" <th>r_hat</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Intercept</th>\n",
" <td>48.199</td>\n",
" <td>8.148</td>\n",
" <td>31.893</td>\n",
" <td>62.836</td>\n",
" <td>0.277</td>\n",
" <td>0.196</td>\n",
" <td>867.0</td>\n",
" <td>887.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>hp</th>\n",
" <td>-0.113</td>\n",
" <td>0.027</td>\n",
" <td>-0.167</td>\n",
" <td>-0.062</td>\n",
" <td>0.001</td>\n",
" <td>0.001</td>\n",
" <td>805.0</td>\n",
" <td>876.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>wt</th>\n",
" <td>-7.838</td>\n",
" <td>1.619</td>\n",
" <td>-10.728</td>\n",
" <td>-4.550</td>\n",
" <td>0.056</td>\n",
" <td>0.040</td>\n",
" <td>827.0</td>\n",
" <td>859.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>hp:wt</th>\n",
" <td>0.026</td>\n",
" <td>0.008</td>\n",
" <td>0.009</td>\n",
" <td>0.041</td>\n",
" <td>0.000</td>\n",
" <td>0.000</td>\n",
" <td>796.0</td>\n",
" <td>694.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>drat</th>\n",
" <td>0.131</td>\n",
" <td>1.217</td>\n",
" <td>-2.316</td>\n",
" <td>2.260</td>\n",
" <td>0.036</td>\n",
" <td>0.029</td>\n",
" <td>1187.0</td>\n",
" <td>1090.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mpg_sigma</th>\n",
" <td>2.308</td>\n",
" <td>0.341</td>\n",
" <td>1.681</td>\n",
" <td>2.934</td>\n",
" <td>0.010</td>\n",
" <td>0.007</td>\n",
" <td>1275.0</td>\n",
" <td>1098.0</td>\n",
" <td>1.0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" mean sd hdi_3% hdi_97% mcse_mean mcse_sd ess_bulk \n",
"Intercept 48.199 8.148 31.893 62.836 0.277 0.196 867.0 \\\n",
"hp -0.113 0.027 -0.167 -0.062 0.001 0.001 805.0 \n",
"wt -7.838 1.619 -10.728 -4.550 0.056 0.040 827.0 \n",
"hp:wt 0.026 0.008 0.009 0.041 0.000 0.000 796.0 \n",
"drat 0.131 1.217 -2.316 2.260 0.036 0.029 1187.0 \n",
"mpg_sigma 2.308 0.341 1.681 2.934 0.010 0.007 1275.0 \n",
"\n",
" ess_tail r_hat \n",
"Intercept 887.0 1.0 \n",
"hp 876.0 1.0 \n",
"wt 859.0 1.0 \n",
"hp:wt 694.0 1.0 \n",
"drat 1090.0 1.0 \n",
"mpg_sigma 1098.0 1.0 "
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"az.summary(idata)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Slopes"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Unit-level slopes"
]
},
{
"cell_type": "code",
"execution_count": 95,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>drat</th>\n",
" <th>wt</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.90</td>\n",
" <td>2.620</td>\n",
" <td>-0.046066</td>\n",
" <td>-0.063662</td>\n",
" <td>-0.029554</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.90</td>\n",
" <td>2.875</td>\n",
" <td>-0.039431</td>\n",
" <td>-0.055364</td>\n",
" <td>-0.024678</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.85</td>\n",
" <td>2.320</td>\n",
" <td>-0.053871</td>\n",
" <td>-0.073766</td>\n",
" <td>-0.033717</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.08</td>\n",
" <td>3.215</td>\n",
" <td>-0.030586</td>\n",
" <td>-0.044648</td>\n",
" <td>-0.014884</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.15</td>\n",
" <td>3.440</td>\n",
" <td>-0.024732</td>\n",
" <td>-0.040846</td>\n",
" <td>-0.009990</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2.76</td>\n",
" <td>3.460</td>\n",
" <td>-0.024211</td>\n",
" <td>-0.039733</td>\n",
" <td>-0.008910</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.21</td>\n",
" <td>3.570</td>\n",
" <td>-0.021350</td>\n",
" <td>-0.037220</td>\n",
" <td>-0.005448</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.69</td>\n",
" <td>3.190</td>\n",
" <td>-0.031236</td>\n",
" <td>-0.046421</td>\n",
" <td>-0.016616</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.92</td>\n",
" <td>3.150</td>\n",
" <td>-0.032277</td>\n",
" <td>-0.047560</td>\n",
" <td>-0.017876</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.92</td>\n",
" <td>3.440</td>\n",
" <td>-0.024732</td>\n",
" <td>-0.040846</td>\n",
" <td>-0.009990</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" term estimate_type drat wt estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx 3.90 2.620 -0.046066 -0.063662 -0.029554\n",
"1 hp dy/dx 3.90 2.875 -0.039431 -0.055364 -0.024678\n",
"2 hp dy/dx 3.85 2.320 -0.053871 -0.073766 -0.033717\n",
"3 hp dy/dx 3.08 3.215 -0.030586 -0.044648 -0.014884\n",
"4 hp dy/dx 3.15 3.440 -0.024732 -0.040846 -0.009990\n",
"5 hp dy/dx 2.76 3.460 -0.024211 -0.039733 -0.008910\n",
"6 hp dy/dx 3.21 3.570 -0.021350 -0.037220 -0.005448\n",
"7 hp dy/dx 3.69 3.190 -0.031236 -0.046421 -0.016616\n",
"8 hp dy/dx 3.92 3.150 -0.032277 -0.047560 -0.017876\n",
"9 hp dy/dx 3.92 3.440 -0.024732 -0.040846 -0.009990"
]
},
"execution_count": 95,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=None\n",
").head(10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Average unit level slopes"
]
},
{
"cell_type": "code",
"execution_count": 96,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>-0.030527</td>\n",
" <td>-0.050973</td>\n",
" <td>-0.010343</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" term estimate_type estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx -0.030527 -0.050973 -0.010343"
]
},
"execution_count": 96,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=None,\n",
" average_by=True\n",
").head(10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Average by subgroup slopes"
]
},
{
"cell_type": "code",
"execution_count": 97,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>drat</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2.76</td>\n",
" <td>-0.023431</td>\n",
" <td>-0.039110</td>\n",
" <td>-0.008132</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2.93</td>\n",
" <td>0.022359</td>\n",
" <td>-0.014317</td>\n",
" <td>0.060480</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.00</td>\n",
" <td>0.026886</td>\n",
" <td>-0.012548</td>\n",
" <td>0.067452</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.07</td>\n",
" <td>-0.013805</td>\n",
" <td>-0.032359</td>\n",
" <td>0.005181</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.08</td>\n",
" <td>-0.022390</td>\n",
" <td>-0.038641</td>\n",
" <td>-0.005317</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.15</td>\n",
" <td>-0.024797</td>\n",
" <td>-0.040907</td>\n",
" <td>-0.010058</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.21</td>\n",
" <td>-0.021350</td>\n",
" <td>-0.037220</td>\n",
" <td>-0.005448</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.23</td>\n",
" <td>0.024830</td>\n",
" <td>-0.013424</td>\n",
" <td>0.064154</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.54</td>\n",
" <td>-0.021350</td>\n",
" <td>-0.037220</td>\n",
" <td>-0.005448</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.62</td>\n",
" <td>-0.042163</td>\n",
" <td>-0.058170</td>\n",
" <td>-0.026350</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" term estimate_type drat estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx 2.76 -0.023431 -0.039110 -0.008132\n",
"1 hp dy/dx 2.93 0.022359 -0.014317 0.060480\n",
"2 hp dy/dx 3.00 0.026886 -0.012548 0.067452\n",
"3 hp dy/dx 3.07 -0.013805 -0.032359 0.005181\n",
"4 hp dy/dx 3.08 -0.022390 -0.038641 -0.005317\n",
"5 hp dy/dx 3.15 -0.024797 -0.040907 -0.010058\n",
"6 hp dy/dx 3.21 -0.021350 -0.037220 -0.005448\n",
"7 hp dy/dx 3.23 0.024830 -0.013424 0.064154\n",
"8 hp dy/dx 3.54 -0.021350 -0.037220 -0.005448\n",
"9 hp dy/dx 3.62 -0.042163 -0.058170 -0.026350"
]
},
"execution_count": 97,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=None,\n",
" average_by=\"drat\"\n",
").head(10)"
]
},
{
"cell_type": "code",
"execution_count": 98,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 700x300 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plot_slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=None,\n",
" average_by=\"wt\"\n",
")\n",
"fig.set_size_inches(7, 3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Elasticity\n",
"\n",
"- Change of slope type from $\\frac{dy}{dx}$ to $\\frac{ey}{ex}$ to get percentage changes"
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>drat</th>\n",
" <th>wt</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.90</td>\n",
" <td>2.620</td>\n",
" <td>-0.219226</td>\n",
" <td>-0.301102</td>\n",
" <td>-0.137880</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.90</td>\n",
" <td>2.875</td>\n",
" <td>-0.198484</td>\n",
" <td>-0.279232</td>\n",
" <td>-0.124738</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.85</td>\n",
" <td>2.320</td>\n",
" <td>-0.196016</td>\n",
" <td>-0.271180</td>\n",
" <td>-0.127203</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.08</td>\n",
" <td>3.215</td>\n",
" <td>-0.167795</td>\n",
" <td>-0.245398</td>\n",
" <td>-0.088351</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3.15</td>\n",
" <td>3.440</td>\n",
" <td>-0.251144</td>\n",
" <td>-0.404628</td>\n",
" <td>-0.082151</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" term estimate_type drat wt estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx 3.90 2.620 -0.219226 -0.301102 -0.137880\n",
"1 hp dy/dx 3.90 2.875 -0.198484 -0.279232 -0.124738\n",
"2 hp dy/dx 3.85 2.320 -0.196016 -0.271180 -0.127203\n",
"3 hp dy/dx 3.08 3.215 -0.167795 -0.245398 -0.088351\n",
"4 hp dy/dx 3.15 3.440 -0.251144 -0.404628 -0.082151"
]
},
"execution_count": 60,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=None,\n",
" slope=\"eyex\"\n",
").head()"
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 700x300 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# plot elasticity\n",
"fig, ax = plot_slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=None,\n",
" slope=\"eyex\",\n",
" average_by=\"wt\"\n",
")\n",
"fig.set_size_inches(7, 3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## User defined slopes\n",
"\n",
"- default and user passed"
]
},
{
"cell_type": "code",
"execution_count": 99,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>wt</th>\n",
" <th>drat</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2</td>\n",
" <td>3.5</td>\n",
" <td>-0.062196</td>\n",
" <td>-0.086427</td>\n",
" <td>-0.038787</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2</td>\n",
" <td>4.0</td>\n",
" <td>-0.062196</td>\n",
" <td>-0.086427</td>\n",
" <td>-0.038787</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2</td>\n",
" <td>4.5</td>\n",
" <td>-0.062196</td>\n",
" <td>-0.086427</td>\n",
" <td>-0.038787</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3</td>\n",
" <td>3.5</td>\n",
" <td>-0.036179</td>\n",
" <td>-0.050728</td>\n",
" <td>-0.020901</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3</td>\n",
" <td>4.0</td>\n",
" <td>-0.036179</td>\n",
" <td>-0.050728</td>\n",
" <td>-0.020901</td>\n",
" </tr>\n",
" <tr>\n",
" <th>10</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3</td>\n",
" <td>4.5</td>\n",
" <td>-0.036179</td>\n",
" <td>-0.050728</td>\n",
" <td>-0.020901</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5</td>\n",
" <td>3.5</td>\n",
" <td>0.015855</td>\n",
" <td>-0.017157</td>\n",
" <td>0.049458</td>\n",
" </tr>\n",
" <tr>\n",
" <th>14</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5</td>\n",
" <td>4.0</td>\n",
" <td>0.015855</td>\n",
" <td>-0.017157</td>\n",
" <td>0.049458</td>\n",
" </tr>\n",
" <tr>\n",
" <th>16</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5</td>\n",
" <td>4.5</td>\n",
" <td>0.015855</td>\n",
" <td>-0.017157</td>\n",
" <td>0.049458</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" term estimate_type wt drat estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx 2 3.5 -0.062196 -0.086427 -0.038787\n",
"2 hp dy/dx 2 4.0 -0.062196 -0.086427 -0.038787\n",
"4 hp dy/dx 2 4.5 -0.062196 -0.086427 -0.038787\n",
"6 hp dy/dx 3 3.5 -0.036179 -0.050728 -0.020901\n",
"8 hp dy/dx 3 4.0 -0.036179 -0.050728 -0.020901\n",
"10 hp dy/dx 3 4.5 -0.036179 -0.050728 -0.020901\n",
"12 hp dy/dx 5 3.5 0.015855 -0.017157 0.049458\n",
"14 hp dy/dx 5 4.0 0.015855 -0.017157 0.049458\n",
"16 hp dy/dx 5 4.5 0.015855 -0.017157 0.049458"
]
},
"execution_count": 99,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# pass your own values for 'wrt' and 'conditional'\n",
"slopes(\n",
" model,\n",
" idata,\n",
" wrt={\"hp\": 50},\n",
" conditional={\"wt\": [2, 3, 5], \"drat\": [3.5, 4, 4.5]}\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 100,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>wt</th>\n",
" <th>drat</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2</td>\n",
" <td>3.5</td>\n",
" <td>-0.062196</td>\n",
" <td>-0.086427</td>\n",
" <td>-0.038787</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2</td>\n",
" <td>4.0</td>\n",
" <td>-0.062196</td>\n",
" <td>-0.086427</td>\n",
" <td>-0.038787</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>2</td>\n",
" <td>4.5</td>\n",
" <td>-0.062196</td>\n",
" <td>-0.086427</td>\n",
" <td>-0.038787</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3</td>\n",
" <td>3.5</td>\n",
" <td>-0.036179</td>\n",
" <td>-0.050728</td>\n",
" <td>-0.020901</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3</td>\n",
" <td>4.0</td>\n",
" <td>-0.036179</td>\n",
" <td>-0.050728</td>\n",
" <td>-0.020901</td>\n",
" </tr>\n",
" <tr>\n",
" <th>10</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>3</td>\n",
" <td>4.5</td>\n",
" <td>-0.036179</td>\n",
" <td>-0.050728</td>\n",
" <td>-0.020901</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5</td>\n",
" <td>3.5</td>\n",
" <td>0.015855</td>\n",
" <td>-0.017157</td>\n",
" <td>0.049458</td>\n",
" </tr>\n",
" <tr>\n",
" <th>14</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5</td>\n",
" <td>4.0</td>\n",
" <td>0.015855</td>\n",
" <td>-0.017157</td>\n",
" <td>0.049458</td>\n",
" </tr>\n",
" <tr>\n",
" <th>16</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5</td>\n",
" <td>4.5</td>\n",
" <td>0.015855</td>\n",
" <td>-0.017157</td>\n",
" <td>0.049458</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" term estimate_type wt drat estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx 2 3.5 -0.062196 -0.086427 -0.038787\n",
"2 hp dy/dx 2 4.0 -0.062196 -0.086427 -0.038787\n",
"4 hp dy/dx 2 4.5 -0.062196 -0.086427 -0.038787\n",
"6 hp dy/dx 3 3.5 -0.036179 -0.050728 -0.020901\n",
"8 hp dy/dx 3 4.0 -0.036179 -0.050728 -0.020901\n",
"10 hp dy/dx 3 4.5 -0.036179 -0.050728 -0.020901\n",
"12 hp dy/dx 5 3.5 0.015855 -0.017157 0.049458\n",
"14 hp dy/dx 5 4.0 0.015855 -0.017157 0.049458\n",
"16 hp dy/dx 5 4.5 0.015855 -0.017157 0.049458"
]
},
"execution_count": 100,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# default 'hp' and user defined for conditional\n",
"slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional={\"wt\": [2, 3, 5], \"drat\": [3.5, 4, 4.5]}\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 101,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 700x300 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plot_slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional={\"wt\": [2, 3, 5], \"drat\": [3.5, 4, 4.5]},\n",
")\n",
"fig.set_size_inches(7, 3)"
]
},
{
"cell_type": "code",
"execution_count": 102,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>term</th>\n",
" <th>estimate_type</th>\n",
" <th>wt</th>\n",
" <th>drat</th>\n",
" <th>estimate</th>\n",
" <th>lower_3.0%</th>\n",
" <th>upper_97.0%</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>1.513</td>\n",
" <td>2.760</td>\n",
" <td>-0.074866</td>\n",
" <td>-0.106387</td>\n",
" <td>-0.045375</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>1.513</td>\n",
" <td>3.080</td>\n",
" <td>-0.074866</td>\n",
" <td>-0.106387</td>\n",
" <td>-0.045375</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>1.513</td>\n",
" <td>3.695</td>\n",
" <td>-0.074866</td>\n",
" <td>-0.106387</td>\n",
" <td>-0.045375</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>1.513</td>\n",
" <td>3.920</td>\n",
" <td>-0.074866</td>\n",
" <td>-0.106387</td>\n",
" <td>-0.045375</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>1.513</td>\n",
" <td>4.930</td>\n",
" <td>-0.074866</td>\n",
" <td>-0.106387</td>\n",
" <td>-0.045375</td>\n",
" </tr>\n",
" <tr>\n",
" <th>...</th>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>490</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5.424</td>\n",
" <td>2.760</td>\n",
" <td>0.026886</td>\n",
" <td>-0.012548</td>\n",
" <td>0.067452</td>\n",
" </tr>\n",
" <tr>\n",
" <th>492</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5.424</td>\n",
" <td>3.080</td>\n",
" <td>0.026886</td>\n",
" <td>-0.012548</td>\n",
" <td>0.067452</td>\n",
" </tr>\n",
" <tr>\n",
" <th>494</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5.424</td>\n",
" <td>3.695</td>\n",
" <td>0.026886</td>\n",
" <td>-0.012548</td>\n",
" <td>0.067452</td>\n",
" </tr>\n",
" <tr>\n",
" <th>496</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5.424</td>\n",
" <td>3.920</td>\n",
" <td>0.026886</td>\n",
" <td>-0.012548</td>\n",
" <td>0.067452</td>\n",
" </tr>\n",
" <tr>\n",
" <th>498</th>\n",
" <td>hp</td>\n",
" <td>dy/dx</td>\n",
" <td>5.424</td>\n",
" <td>4.930</td>\n",
" <td>0.026886</td>\n",
" <td>-0.012548</td>\n",
" <td>0.067452</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>250 rows × 7 columns</p>\n",
"</div>"
],
"text/plain": [
" term estimate_type wt drat estimate lower_3.0% upper_97.0%\n",
"0 hp dy/dx 1.513 2.760 -0.074866 -0.106387 -0.045375\n",
"2 hp dy/dx 1.513 3.080 -0.074866 -0.106387 -0.045375\n",
"4 hp dy/dx 1.513 3.695 -0.074866 -0.106387 -0.045375\n",
"6 hp dy/dx 1.513 3.920 -0.074866 -0.106387 -0.045375\n",
"8 hp dy/dx 1.513 4.930 -0.074866 -0.106387 -0.045375\n",
".. ... ... ... ... ... ... ...\n",
"490 hp dy/dx 5.424 2.760 0.026886 -0.012548 0.067452\n",
"492 hp dy/dx 5.424 3.080 0.026886 -0.012548 0.067452\n",
"494 hp dy/dx 5.424 3.695 0.026886 -0.012548 0.067452\n",
"496 hp dy/dx 5.424 3.920 0.026886 -0.012548 0.067452\n",
"498 hp dy/dx 5.424 4.930 0.026886 -0.012548 0.067452\n",
"\n",
"[250 rows x 7 columns]"
]
},
"execution_count": 102,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# default 'wrt' and 'conditional'\n",
"slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=[\"wt\", \"drat\"]\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 66,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 1200x500 with 6 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig, ax = plot_slopes(\n",
" model,\n",
" idata,\n",
" wrt=\"hp\",\n",
" conditional=[\"wt\", \"drat\"],\n",
" subplot_kwargs={\"main\": \"wt\", \"group\": \"drat\", \"panel\": \"drat\"},\n",
" fig_kwargs={\"sharey\": True, \"figsize\": (12, 5)},\n",
" legend=False\n",
")\n",
"plt.tight_layout()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "bambinos",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.11.0"
},
"orig_nbformat": 4
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment