Merge branch 'master' of https://github.com/NBISweden/workshop-mlbios…

…tatistics

session-lm/.quarto/cites/index.json

@@ -1 +1 @@
-{"lm-lasso-exercises.qmd":[],"lm-intro-exercises.qmd":[],"lm-intro.qmd":[],"lm-diagn-exercises.qmd":[],"lm-diagn.qmd":[],"lm-glm.qmd":[],"lm-reg-cls.qmd":[],"lm-coeff-exercises.qmd":[],"lm-coeff.qmd":[],"index.qmd":[],"lm-ml.qmd":[]}
+{"lm-ml.qmd":[],"lm-lasso-exercises.qmd":[],"lm-diagn.qmd":[],"lm-coeff.qmd":[],"lm-intro.qmd":[],"lm-diagn-exercises.qmd":[],"lm-glm.qmd":[],"lm-intro-exercises.qmd":[],"lm-reg-cls.qmd":[],"index.qmd":[],"lm-coeff-exercises.qmd":[]}

session-lm/.quarto/xref/28d59191

@@ -1 +1 @@
-{"entries":[],"options":{"chapters":true},"headings":["preface"]}
+{"options":{"chapters":true},"entries":[],"headings":["preface"]}

session-lm/.quarto/xref/295884aa

@@ -1 +1 @@
-{"headings":["fitting-linear-model","hypothesis-testing","evaluate-model-fit"],"entries":[{"key":"exr-lm-assess-fit","caption":"Evaluate model fit","order":{"number":3,"section":[0,0,0,0,0,0,0]}},{"key":"exr-lm-fit","caption":"Fitting linear model","order":{"number":1,"section":[0,0,0,0,0,0,0]}},{"key":"exr-lm-hypothesis","caption":"Hypothesis testing","order":{"number":2,"section":[0,0,0,0,0,0,0]}}],"options":{"chapters":true}}
+{"entries":[{"key":"exr-lm-fit","caption":"Fitting linear model","order":{"number":1,"section":[0,0,0,0,0,0,0]}},{"key":"exr-lm-hypothesis","caption":"Hypothesis testing","order":{"number":2,"section":[0,0,0,0,0,0,0]}},{"key":"exr-lm-assess-fit","caption":"Evaluate model fit","order":{"number":3,"section":[0,0,0,0,0,0,0]}}],"headings":["fitting-linear-model","hypothesis-testing","evaluate-model-fit"],"options":{"chapters":true}}

session-lm/.quarto/xref/4d506fc9

@@ -1 +1 @@
-{"entries":[{"caption":"Brozek score","key":"exr-brozek","order":{"section":[0,0,0,0,0,0,0],"number":1}}],"options":{"chapters":true},"headings":["brozek-score","answers-to-selected-exercises"]}
+{"headings":["brozek-score","answers-to-selected-exercises"],"entries":[{"caption":"Brozek score","key":"exr-brozek","order":{"section":[0,0,0,0,0,0,0],"number":1}}],"options":{"chapters":true}}

session-lm/.quarto/xref/5b28360f

@@ -1 +1 @@
-{"entries":[{"order":{"number":2,"section":[0,0,0,0,0,0,0]},"caption":"Trout","key":"exr-trout"},{"order":{"number":1,"section":[0,0,0,0,0,0,0]},"caption":"Height-weight-gender","key":"exr-rerun"},{"order":{"number":3,"section":[0,0,0,0,0,0,0]},"caption":"Lowering blood pressure","key":"exr-drug"}],"options":{"chapters":true},"headings":["height-weight-gender","trout","lowering-blood-pressure","answers-to-selected-exercises"]}
+{"headings":["height-weight-gender","trout","lowering-blood-pressure","answers-to-selected-exercises"],"entries":[{"caption":"Trout","key":"exr-trout","order":{"section":[0,0,0,0,0,0,0],"number":2}},{"caption":"Height-weight-gender","key":"exr-rerun","order":{"section":[0,0,0,0,0,0,0],"number":1}},{"caption":"Lowering blood pressure","key":"exr-drug","order":{"section":[0,0,0,0,0,0,0],"number":3}}],"options":{"chapters":true}}

session-lm/.quarto/xref/691de0dc

@@ -1 +1 @@
-{"options":{"chapters":true},"entries":[{"order":{"number":1,"section":[3,1,0,0,0,0,0]},"caption":"Scatter plot showing BMI values given waist measurments with a fitted simple linear regression model.","key":"fig-simple-1"}],"headings":["example-simple-linear-regression","example-multiple-regression","example-categorical-variable","example-categorical-numerical-variables","example-interactions","example-logistic-regression-with-categorical-variable"]}
+{"options":{"chapters":true},"headings":["example-simple-linear-regression","example-multiple-regression","example-categorical-variable","example-categorical-numerical-variables","example-interactions","example-logistic-regression-with-categorical-variable"],"entries":[{"key":"fig-simple-1","caption":"Scatter plot showing BMI values given waist measurments with a fitted simple linear regression model.","order":{"number":1,"section":[3,1,0,0,0,0,0]}}]}

session-lm/.quarto/xref/9e3bf041

@@ -1 +1 @@
-{"options":{"chapters":true},"entries":[{"caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca. Linear regression gives the equation of the straight line (red) that best describes how the outcome changes (increase or decreases) with a change of exposure variable","key":"fig-lm-example-reg","order":{"section":[1,5,0,0,0,0,0],"number":5}},{"caption":"vector-matrix-notation","key":"exm-vector-matrix-notation","order":{"section":[1,9,0,0,0,0,0],"number":4}},{"caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca. Linear regrssion gives the equation of the straight line (red) that best describes how the outcome changes with a change of exposure variable. Blue lines represent error terms, the vertical distances to the regression line","key":"fig-reg-errors","order":{"section":[1,6,0,0,0,0,0],"number":6}},{"caption":"Prediction and intervals","key":"exm-prediction-and-intervals","order":{"section":[1,10,0,0,0,0,0],"number":5}},{"caption":"Least squares in vector-matrix notation","key":"thm-lss-vector-matrix","order":{"section":[1,9,0,0,0,0,0],"number":2}},{"caption":"vector matrix form of the linear model","key":"def-vector-matrix-lm","order":{"section":[1,9,0,0,0,0,0],"number":1}},{"caption":"Hypothesis testing","key":"exm-hypothesis-testing","order":{"section":[1,8,0,0,0,0,0],"number":3}},{"caption":"Scatter plot of weight vs. height for the 130 study participants based on the diabetes data set collected to understand the prevalence of obesity, diabetes, and other cardiovascular risk factors in central Virginia, USA.","key":"fig-scatter","order":{"section":[1,1,0,0,0,0,0],"number":1}},{"caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca.","key":"fig-lm-intro-example","order":{"section":[1,5,0,0,0,0,0],"number":4}},{"caption":"Least squares","key":"exm-lss","order":{"section":[1,6,0,0,0,0,0],"number":2}},{"caption":"Weight and plasma volume","key":"exm-simple-lm","order":{"section":[1,5,0,0,0,0,0],"number":1}},{"caption":"Deterministic vs. statistical relationship: a) deterministic: equation exactly describes the relationship between the two variables e.g. Ferenheit and Celcius relationship, b) statistical relationship between x and y is not perfect (increasing relationship), c) statistical relationship between x and y is not perfect (decreasing relationship), d) random signal","key":"fig-relationship","order":{"section":[1,2,0,0,0,0,0],"number":2}},{"caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca. Linear regression gives the equation of the straight line that best describes how the outcome changes (increase or decreases) with a change of exposure variable (in red)","key":"fig-lm-parameters","order":{"section":[1,7,0,0,0,0,0],"number":7}},{"caption":"R^2(adj)","key":"thm-r2adj","order":{"section":[1,11,0,0,0,0,0],"number":4}},{"caption":"R^2","key":"def-r2","order":{"section":[1,11,0,0,0,0,0],"number":2}},{"caption":"Least squares estimates for a simple linear regression","key":"thm-lss","order":{"section":[1,6,0,0,0,0,0],"number":1}},{"caption":"R^2","key":"thm-r2","order":{"section":[1,11,0,0,0,0,0],"number":3}},{"caption":"Examples of a linear models: A) y_i = x_1 + e_i, B) x_1 + I_{x_i} + e_i C) y_i = x_i^2 + e_i, D) y_i = x + x_i^3 + e_i showing that linear models can get more complex and/or capture more than a straight line relationship.","key":"fig-linear-adv","order":{"section":[1,3,0,0,0,0,0],"number":3}},{"caption":"","key":"eq-lm","order":{"section":[1,5,0,0,0,0,0],"number":1}}],"headings":["why-linear-models","statistical-vs.-deterministic-relationship","what-linear-models-are-and-are-not","terminology","simple-linear-regression","weight-and-plasma-volume","least-squares","least-squares-estimates-for-a-simple-linear-regression","least-squares-1","intercept-and-slope","hypothesis-testing","hypothesis-testing-1","vector-matrix-notations","vector-matrix-form-of-the-linear-model","least-squares-in-vector-matrix-notation","vector-matrix-notation","confidence-intervals-and-prediction-intervals","prediction-and-intervals","assessing-model-fit","r2-summary-of-the-fitted-model","r2","r2-and-correlation-coefficient","r2-1","r2adj","r2adj-1","the-assumptions-of-a-linear-model","checking-assumptions","influential-observations"]}
+{"options":{"chapters":true},"headings":["why-linear-models","statistical-vs.-deterministic-relationship","what-linear-models-are-and-are-not","terminology","simple-linear-regression","weight-and-plasma-volume","least-squares","least-squares-estimates-for-a-simple-linear-regression","least-squares-1","intercept-and-slope","hypothesis-testing","hypothesis-testing-1","vector-matrix-notations","vector-matrix-form-of-the-linear-model","least-squares-in-vector-matrix-notation","vector-matrix-notation","confidence-intervals-and-prediction-intervals","prediction-and-intervals","assessing-model-fit","r2-summary-of-the-fitted-model","r2","r2-and-correlation-coefficient","r2-1","r2adj","r2adj-1","the-assumptions-of-a-linear-model","checking-assumptions","influential-observations"],"entries":[{"key":"def-r2","caption":"R^2","order":{"section":[1,11,0,0,0,0,0],"number":2}},{"key":"eq-lm","caption":"","order":{"section":[1,5,0,0,0,0,0],"number":1}},{"key":"exm-lss","caption":"Least squares","order":{"section":[1,6,0,0,0,0,0],"number":2}},{"key":"thm-lss","caption":"Least squares estimates for a simple linear regression","order":{"section":[1,6,0,0,0,0,0],"number":1}},{"key":"fig-scatter","caption":"Scatter plot of weight vs. height for the 130 study participants based on the diabetes data set collected to understand the prevalence of obesity, diabetes, and other cardiovascular risk factors in central Virginia, USA.","order":{"section":[1,1,0,0,0,0,0],"number":1}},{"key":"fig-linear-adv","caption":"Examples of a linear models: A) y_i = x_1 + e_i, B) x_1 + I_{x_i} + e_i C) y_i = x_i^2 + e_i, D) y_i = x + x_i^3 + e_i showing that linear models can get more complex and/or capture more than a straight line relationship.","order":{"section":[1,3,0,0,0,0,0],"number":3}},{"key":"exm-vector-matrix-notation","caption":"vector-matrix-notation","order":{"section":[1,9,0,0,0,0,0],"number":4}},{"key":"thm-r2adj","caption":"R^2(adj)","order":{"section":[1,11,0,0,0,0,0],"number":4}},{"key":"exm-prediction-and-intervals","caption":"Prediction and intervals","order":{"section":[1,10,0,0,0,0,0],"number":5}},{"key":"thm-r2","caption":"R^2","order":{"section":[1,11,0,0,0,0,0],"number":3}},{"key":"fig-reg-errors","caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca. Linear regrssion gives the equation of the straight line (red) that best describes how the outcome changes with a change of exposure variable. Blue lines represent error terms, the vertical distances to the regression line","order":{"section":[1,6,0,0,0,0,0],"number":6}},{"key":"exm-simple-lm","caption":"Weight and plasma volume","order":{"section":[1,5,0,0,0,0,0],"number":1}},{"key":"fig-lm-parameters","caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca. Linear regression gives the equation of the straight line that best describes how the outcome changes (increase or decreases) with a change of exposure variable (in red)","order":{"section":[1,7,0,0,0,0,0],"number":7}},{"key":"thm-lss-vector-matrix","caption":"Least squares in vector-matrix notation","order":{"section":[1,9,0,0,0,0,0],"number":2}},{"key":"def-vector-matrix-lm","caption":"vector matrix form of the linear model","order":{"section":[1,9,0,0,0,0,0],"number":1}},{"key":"fig-lm-example-reg","caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca. Linear regression gives the equation of the straight line (red) that best describes how the outcome changes (increase or decreases) with a change of exposure variable","order":{"section":[1,5,0,0,0,0,0],"number":5}},{"key":"fig-lm-intro-example","caption":"Scatter plot of the data shows that high plasma volume tends to be associated with high weight and vice verca.","order":{"section":[1,5,0,0,0,0,0],"number":4}},{"key":"fig-relationship","caption":"Deterministic vs. statistical relationship: a) deterministic: equation exactly describes the relationship between the two variables e.g. Ferenheit and Celcius relationship, b) statistical relationship between x and y is not perfect (increasing relationship), c) statistical relationship between x and y is not perfect (decreasing relationship), d) random signal","order":{"section":[1,2,0,0,0,0,0],"number":2}},{"key":"exm-hypothesis-testing","caption":"Hypothesis testing","order":{"section":[1,8,0,0,0,0,0],"number":3}}]}

session-lm/docs/lm-coeff.html

@@ -442,7 +442,7 @@ <h2 data-number="3.2" class="anchored" data-anchor-id="example-multiple-regressi
 </ul>
 <p><strong>Specific interpretation</strong></p>
 <ul>
-<li>Obviously there is difference between decrease of 0.9 BMI and decrease of 0.9 in BMI (alternative model).</li>
+<li>Obviously there is difference between decrease of 0.9 BMI and decrease of 0.06 in BMI (alternative model).</li>
 <li>Our interpretations need to be more specific and we say that <strong>a unit increase in <span class="math inline">\(x\)</span> with other predictors held constant will produce a change equal to <span class="math inline">\(\hat{\beta}\)</span> in the response <span class="math inline">\(y\)</span></strong></li>
 <li>Often it may be quite unrealistic to be able to control other variables and keep them constant and for our alternative model, a change in <code>hdl</code> would also imply a change in total cholesterol <code>chol</code>.</li>
 <li>Further, our explanation contains <strong>no notation of causation</strong>.</li>
@@ -590,7 +590,7 @@ <h2 data-number="3.4" class="anchored" data-anchor-id="example-categorical-numer
         \end{array}
     \right.
 \end{equation}\]</span></p>
-<p>and <span class="math inline">\(x_{2,i}\)</span> is the <code>heigth</code> of person <span class="math inline">\(i\)</span>.</p>
+<p>and <span class="math inline">\(x_{2,i}\)</span> is the <code>height</code> of person <span class="math inline">\(i\)</span>.</p>
 <p>In <code>R</code> we write:</p>
 <div class="cell">
 <div class="sourceCode cell-code" id="cb12"><pre class="sourceCode r code-with-copy"><code class="sourceCode r"><span id="cb12-1"><a href="#cb12-1" aria-hidden="true" tabindex="-1"></a><span class="co"># fit linear model and print model summary</span></span>