Done inference

brycesisu · Dec 5, 2017 · 9c02ca4 · 9c02ca4
1 parent e2950fd
commit 9c02ca4
Show file tree

Hide file tree

Showing 15 changed files with 157 additions and 80 deletions.
diff --git a/06-regression.Rmd b/06-regression.Rmd
@@ -1387,3 +1387,36 @@ lm(score ~ age * gender, data=evals) %>%
     booktabs = TRUE
   )
 ```
+
+
+
+### Refresher: Residual analysis
+
+Let's compute the residuals using `augment()` to see if there is a pattern.
+
+```{r residual0, warning=FALSE}
+point_by_point_info <- lm(score ~ age * gender, data=evals) %>% 
+  augment() %>% 
+  select(score, age, gender, .fitted, .resid)
+```
+
+First the histogram:
+
+```{r residual1, warning=FALSE, fig.cap="Model 2 (with interaction) histogram of residual"}
+ggplot(point_by_point_info, aes(x=.resid)) +
+  geom_histogram(binwidth = 0.25) +
+  labs(x="Residual")
+```
+
+Second, the residuals as compared to the predictor variables:
+
+* $x_1$: numerical explanatory/predictor variable of `age`
+* $x_2$: categorical explanatory/predictor variable of `gender`
+
+```{r residual2, warning=FALSE, fig.cap="Model 2 (with interaction) residuals vs predictor"}
+ggplot(point_by_point_info, aes(x=age, y=.resid)) +
+  geom_point() + 
+  labs(x="age", y="Residual") +
+  geom_hline(yintercept=0, col="blue", size =1) +
+  facet_wrap(~gender)
+```
diff --git a/docs/2-getting-started.html b/docs/2-getting-started.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/3-viz.html b/docs/3-viz.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/4-tidy.html b/docs/4-tidy.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/5-wrangling.html b/docs/5-wrangling.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/6-regression.html b/docs/6-regression.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>
@@ -437,53 +438,53 @@ <h3><span class="header-section-number">6.1.1</span> Professor evaluations</h3>
 </thead>
 <tbody>
 <tr class="odd">
-<td>290</td>
-<td align="right">3.6</td>
-<td align="left">not minority</td>
+<td>349</td>
+<td align="right">5.0</td>
+<td align="left">minority</td>
 <td align="left">male</td>
 <td align="left">english</td>
-<td align="right">34</td>
-<td align="right">6.67</td>
-<td align="left">tenure track</td>
+<td align="right">50</td>
+<td align="right">3.33</td>
+<td align="left">teaching</td>
 </tr>
 <tr class="even">
-<td>350</td>
-<td align="right">4.9</td>
-<td align="left">minority</td>
+<td>397</td>
+<td align="right">4.8</td>
+<td align="left">not minority</td>
 <td align="left">male</td>
 <td align="left">english</td>
-<td align="right">50</td>
+<td align="right">45</td>
 <td align="right">3.33</td>
 <td align="left">teaching</td>
 </tr>
 <tr class="odd">
-<td>84</td>
-<td align="right">4.2</td>
+<td>165</td>
+<td align="right">3.6</td>
 <td align="left">not minority</td>
 <td align="left">male</td>
 <td align="left">english</td>
-<td align="right">45</td>
-<td align="right">4.17</td>
-<td align="left">tenured</td>
+<td align="right">63</td>
+<td align="right">4.33</td>
+<td align="left">teaching</td>
 </tr>
 <tr class="even">
-<td>62</td>
-<td align="right">4.3</td>
-<td align="left">minority</td>
-<td align="left">male</td>
+<td>451</td>
+<td align="right">4.4</td>
+<td align="left">not minority</td>
+<td align="left">female</td>
 <td align="left">non-english</td>
-<td align="right">35</td>
-<td align="right">4.83</td>
+<td align="right">60</td>
+<td align="right">4.33</td>
 <td align="left">tenure track</td>
 </tr>
 <tr class="odd">
-<td>277</td>
-<td align="right">4.9</td>
+<td>143</td>
+<td align="right">4.3</td>
 <td align="left">not minority</td>
-<td align="left">female</td>
+<td align="left">male</td>
 <td align="left">english</td>
-<td align="right">38</td>
-<td align="right">6.50</td>
+<td align="right">58</td>
+<td align="right">3.83</td>
 <td align="left">tenured</td>
 </tr>
 </tbody>
@@ -517,34 +518,34 @@ <h3><span class="header-section-number">6.1.2</span> Country-by-country life exp
 </thead>
 <tbody>
 <tr class="odd">
-<td align="left">Spain</td>
-<td align="right">80.9</td>
-<td align="left">Europe</td>
-<td align="right">28821.1</td>
+<td align="left">Kenya</td>
+<td align="right">54.1</td>
+<td align="left">Africa</td>
+<td align="right">1463.2</td>
 </tr>
 <tr class="even">
-<td align="left">Congo, Dem. Rep.</td>
-<td align="right">46.5</td>
+<td align="left">Gambia</td>
+<td align="right">59.4</td>
 <td align="left">Africa</td>
-<td align="right">277.6</td>
+<td align="right">752.7</td>
 </tr>
 <tr class="odd">
-<td align="left">Turkey</td>
-<td align="right">71.8</td>
-<td align="left">Europe</td>
-<td align="right">8458.3</td>
+<td align="left">Benin</td>
+<td align="right">56.7</td>
+<td align="left">Africa</td>
+<td align="right">1441.3</td>
 </tr>
 <tr class="even">
-<td align="left">Sri Lanka</td>
-<td align="right">72.4</td>
-<td align="left">Asia</td>
-<td align="right">3970.1</td>
+<td align="left">Norway</td>
+<td align="right">80.2</td>
+<td align="left">Europe</td>
+<td align="right">49357.2</td>
 </tr>
 <tr class="odd">
-<td align="left">Singapore</td>
-<td align="right">80.0</td>
-<td align="left">Asia</td>
-<td align="right">47143.2</td>
+<td align="left">Congo, Dem. Rep.</td>
+<td align="right">46.5</td>
+<td align="left">Africa</td>
+<td align="right">277.6</td>
 </tr>
 </tbody>
 </table>
@@ -576,34 +577,34 @@ <h3><span class="header-section-number">6.1.3</span> Credit card balance</h3>
 </thead>
 <tbody>
 <tr class="odd">
-<td>136</td>
-<td align="right">8</td>
-<td align="right">44.5</td>
-<td align="right">3500</td>
+<td>382</td>
+<td align="right">849</td>
+<td align="right">101.8</td>
+<td align="right">8029</td>
 </tr>
 <tr class="even">
-<td>268</td>
-<td align="right">942</td>
-<td align="right">29.6</td>
-<td align="right">5833</td>
+<td>289</td>
+<td align="right">863</td>
+<td align="right">16.7</td>
+<td align="right">5274</td>
 </tr>
 <tr class="odd">
-<td>44</td>
-<td align="right">976</td>
-<td align="right">36.9</td>
-<td align="right">6257</td>
+<td>246</td>
+<td align="right">0</td>
+<td align="right">34.5</td>
+<td align="right">2001</td>
 </tr>
 <tr class="even">
-<td>116</td>
-<td align="right">510</td>
-<td align="right">40.4</td>
-<td align="right">4828</td>
+<td>257</td>
+<td align="right">0</td>
+<td align="right">25.9</td>
+<td align="right">1774</td>
 </tr>
 <tr class="odd">
-<td>22</td>
-<td align="right">968</td>
-<td align="right">37.3</td>
-<td align="right">6378</td>
+<td>126</td>
+<td align="right">0</td>
+<td align="right">27.6</td>
+<td align="right">2531</td>
 </tr>
 </tbody>
 </table>
@@ -628,33 +629,33 @@ <h3><span class="header-section-number">6.1.4</span> Biopics</h3>
 </thead>
 <tbody>
 <tr class="odd">
-<td align="left">Midnight Express</td>
-<td align="right">35000000</td>
-<td align="left">FALSE</td>
+<td align="left">Invictus</td>
+<td align="right">37500000</td>
+<td align="left">TRUE</td>
 <td align="left">Male</td>
 </tr>
 <tr class="even">
-<td align="left">127 Hours</td>
-<td align="right">18300000</td>
-<td align="left">FALSE</td>
+<td align="left">Music Within</td>
+<td align="right">153000</td>
+<td align="left">TRUE</td>
 <td align="left">Male</td>
 </tr>
 <tr class="odd">
-<td align="left">The Lover</td>
-<td align="right">4900000</td>
+<td align="left">Dallas Buyers Club</td>
+<td align="right">27300000</td>
 <td align="left">FALSE</td>
-<td align="left">Female</td>
+<td align="left">Male</td>
 </tr>
 <tr class="even">
-<td align="left">Rudy</td>
-<td align="right">22800000</td>
+<td align="left">Catch Me If You Can</td>
+<td align="right">164000000</td>
 <td align="left">FALSE</td>
 <td align="left">Male</td>
 </tr>
 <tr class="odd">
-<td align="left">My Left Foot</td>
-<td align="right">14700000</td>
-<td align="left">FALSE</td>
+<td align="left">Selma</td>
+<td align="right">52100000</td>
+<td align="left">TRUE</td>
 <td align="left">Male</td>
 </tr>
 </tbody>
@@ -2108,6 +2109,39 @@ <h3><span class="header-section-number">6.7.4</span> New output: Confidence inte
 </tr>
 </tbody>
 </table>
+</div>
+<div id="refresher-residual-analysis" class="section level3">
+<h3><span class="header-section-number">6.7.5</span> Refresher: Residual analysis</h3>
+<p>Let’s compute the residuals using <code>augment()</code> to see if there is a pattern.</p>
+<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r">point_by_point_info &lt;-<span class="st"> </span><span class="kw">lm</span>(score <span class="op">~</span><span class="st"> </span>age <span class="op">*</span><span class="st"> </span>gender, <span class="dt">data=</span>evals) <span class="op">%&gt;%</span><span class="st"> </span>
+<span class="st">  </span><span class="kw">augment</span>() <span class="op">%&gt;%</span><span class="st"> </span>
+<span class="st">  </span><span class="kw">select</span>(score, age, gender, .fitted, .resid)</code></pre></div>
+<p>First the histogram:</p>
+<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">ggplot</span>(point_by_point_info, <span class="kw">aes</span>(<span class="dt">x=</span>.resid)) <span class="op">+</span>
+<span class="st">  </span><span class="kw">geom_histogram</span>(<span class="dt">binwidth =</span> <span class="fl">0.25</span>) <span class="op">+</span>
+<span class="st">  </span><span class="kw">labs</span>(<span class="dt">x=</span><span class="st">&quot;Residual&quot;</span>)</code></pre></div>
+<div class="figure" style="text-align: center"><span id="fig:residual1"></span>
+<img src="ismaykim_files/figure-html/residual1-1.png" alt="Model 2 (with interaction) histogram of residual" width="\textwidth" />
+<p class="caption">
+Figure 6.27: Model 2 (with interaction) histogram of residual
+</p>
+</div>
+<p>Second, the residuals as compared to the predictor variables:</p>
+<ul>
+<li><span class="math inline">\(x_1\)</span>: numerical explanatory/predictor variable of <code>age</code></li>
+<li><span class="math inline">\(x_2\)</span>: categorical explanatory/predictor variable of <code>gender</code></li>
+</ul>
+<div class="sourceCode"><pre class="sourceCode r"><code class="sourceCode r"><span class="kw">ggplot</span>(point_by_point_info, <span class="kw">aes</span>(<span class="dt">x=</span>age, <span class="dt">y=</span>.resid)) <span class="op">+</span>
+<span class="st">  </span><span class="kw">geom_point</span>() <span class="op">+</span><span class="st"> </span>
+<span class="st">  </span><span class="kw">labs</span>(<span class="dt">x=</span><span class="st">&quot;age&quot;</span>, <span class="dt">y=</span><span class="st">&quot;Residual&quot;</span>) <span class="op">+</span>
+<span class="st">  </span><span class="kw">geom_hline</span>(<span class="dt">yintercept=</span><span class="dv">0</span>, <span class="dt">col=</span><span class="st">&quot;blue&quot;</span>, <span class="dt">size =</span><span class="dv">1</span>) <span class="op">+</span>
+<span class="st">  </span><span class="kw">facet_wrap</span>(<span class="op">~</span>gender)</code></pre></div>
+<div class="figure" style="text-align: center"><span id="fig:residual2"></span>
+<img src="ismaykim_files/figure-html/residual2-1.png" alt="Model 2 (with interaction) residuals vs predictor" width="\textwidth" />
+<p class="caption">
+Figure 6.28: Model 2 (with interaction) residuals vs predictor
+</p>
+</div>
 
 </div>
 </div>

diff --git a/docs/7-sampling.html b/docs/7-sampling.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/A-appendixA.html b/docs/A-appendixA.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/B-appendixB.html b/docs/B-appendixB.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/C-appendixC.html b/docs/C-appendixC.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/index.html b/docs/index.html
@@ -278,6 +278,7 @@
 <li class="chapter" data-level="6.7.2" data-path="6-regression.html"><a href="6-regression.html#refresher-visualizations"><i class="fa fa-check"></i><b>6.7.2</b> Refresher: Visualizations</a></li>
 <li class="chapter" data-level="6.7.3" data-path="6-regression.html"><a href="6-regression.html#refresher-regression-tables"><i class="fa fa-check"></i><b>6.7.3</b> Refresher: Regression tables</a></li>
 <li class="chapter" data-level="6.7.4" data-path="6-regression.html"><a href="6-regression.html#new-output-confidence-intervals"><i class="fa fa-check"></i><b>6.7.4</b> New output: Confidence intervals</a></li>
+<li class="chapter" data-level="6.7.5" data-path="6-regression.html"><a href="6-regression.html#refresher-residual-analysis"><i class="fa fa-check"></i><b>6.7.5</b> Refresher: Residual analysis</a></li>
 </ul></li>
 </ul></li>
 <li class="part"><span><b>II Inference</b></span></li>

diff --git a/docs/ismaykim_files/figure-html/jitter-1.png b/docs/ismaykim_files/figure-html/jitter-1.png
diff --git a/docs/ismaykim_files/figure-html/numxplot2-1.png b/docs/ismaykim_files/figure-html/numxplot2-1.png