Decision Trees and Ensemble Learning Quiz

<p>Creating multiple decision trees each trained on a different bootstrap sample of the data</p> Signup and view all the answers

<p>$1 - \sum p_i^2$</p> Signup and view all the answers

<p>Higher statistical significance of differences between sub-node and parent node</p> Signup and view all the answers

<p>$-\sum p_i \log_2(p_i)$</p> Signup and view all the answers

<p>Starting at the bottom and removing leaves with negative returns compared to the top</p> Signup and view all the answers

<p>To limit the number of features considered while searching for a best split</p> Signup and view all the answers

<p>A less impure node requires less information to describe it</p> Signup and view all the answers

<p>Higher homogeneity within each subset after splitting</p> Signup and view all the answers

<p>The spread or dispersion of data within each node</p> Signup and view all the answers

<p>Overfitting by creating an overly complex model</p> Signup and view all the answers

<p>Final prediction outcomes or subsets at the end of branches</p> Signup and view all the answers

<p>Sampling with replacement, where some data is left out of each tree in the sample</p> Signup and view all the answers

<p>A bag of decision trees using subspace sampling</p> Signup and view all the answers

<p>Aggregates weak learners to form a strong predictor by adding new trees that minimize the error of previous learners</p> Signup and view all the answers

<p>By following decisions in the tree from the root to a leaf node, using branching conditions and trained weights</p> Signup and view all the answers

<p>Bagged decision trees consist of independently trained trees on bootstrapped data, while boosting adds weak learners iteratively</p> Signup and view all the answers

<p>Classification or regression trees</p> Signup and view all the answers

<p>Regression trees predict continuous values, while classification trees represent class labels on leaves and conjunctions of features on branches.</p> Signup and view all the answers

<p>A loss function, choosing the best variable and splitting value among all possibilities.</p> Signup and view all the answers

<p>Criteria like maximum depth, node size, and pruning.</p> Signup and view all the answers

<p>In a financial market context.</p> Signup and view all the answers

<p>Ensembles of random trees, like bootstrapping with decision trees</p> Signup and view all the answers

<p>Shannon's Entropy Model</p> Signup and view all the answers

<p>Pruning</p> Signup and view all the answers

<p>Aggregating the results of different models</p> Signup and view all the answers

<p>Creating multiple decision trees each trained on a different bootstrap sample of the data</p> Signup and view all the answers

<p>To reduce overfitting by limiting tree depth</p> Signup and view all the answers

<p>They are turned into categorical variables before a split at the root node</p> Signup and view all the answers

<p>Bagging involves creating multiple decision trees trained on different bootstrap samples, while boosting adapts the weights of data points at each iteration</p> Signup and view all the answers

<p>Impurity to result in a set that minimizes impurity</p> Signup and view all the answers

<p>Higher impurity within the split</p> Signup and view all the answers

<p>Predicted outcomes</p> Signup and view all the answers

<p>$G = 1 - \sum_{i=1}^{n} p_i^2$</p> Signup and view all the answers

<p>To limit the number of features considered for each split</p> Signup and view all the answers

<p>Less impure node requires less information to describe it</p> Signup and view all the answers

<p>Higher statistical significance between sub-nodes and parent node</p> Signup and view all the answers

<p>Overfitting</p> Signup and view all the answers

<p>Impurity</p> Signup and view all the answers

<p>Bagging aims to reduce variance while boosting aims to reduce bias.</p> Signup and view all the answers

<p>Pruning techniques</p> Signup and view all the answers

<p>It sequentially trains multiple models to correct errors made by previous models.</p> Signup and view all the answers

<p>$Gini = 1 - igg(\sum_{i=1}^{n} P_i^2\bigg)$</p> Signup and view all the answers

<p>To prevent overfitting by reducing the size of the tree</p> Signup and view all the answers

<p>The final predicted outcome or classification</p> Signup and view all the answers

<p>Higher statistical significance of differences between sub-node and parent node</p> Signup and view all the answers

<p>Bagging builds multiple models independently, while boosting builds models sequentially to correct errors</p> Signup and view all the answers

<p>Variance</p> Signup and view all the answers

<p>Gini Index</p> Signup and view all the answers

<p>To reduce both bias and variance simultaneously</p> Signup and view all the answers

<p>Less impure node requiring less information to describe it</p> Signup and view all the answers

<p>By creating binary splits based on threshold values</p> Signup and view all the answers

<p>Regression trees predict continuous outcomes, while classification trees predict categorical outcomes</p> Signup and view all the answers

<p>Bagged decision trees consist of independently trained trees on bootstrapped data, while boosting adds weak learners iteratively.</p> Signup and view all the answers

<p>To ensure interpretability and prevent overfitting in decision trees.</p> Signup and view all the answers

<p>As independent variables in a financial market context.</p> Signup and view all the answers

<p>A loss function is minimized to fit a decision tree, choosing the best variable and splitting value among all possibilities.</p> Signup and view all the answers

<p>Overfitting in decision trees.</p> Signup and view all the answers

<p>Bootstrapping involves sampling with replacement, where some data is left out of each tree in the sample.</p> Signup and view all the answers

<p>Regression trees predict continuous values, while classification trees represent class labels on leaves and conjunctions of features on branches.</p> Signup and view all the answers

<p>Higher homogeneity in the split.</p> Signup and view all the answers

<p>CART produces classification or regression trees, depending on the dependent variable's nature.</p> Signup and view all the answers

<p>Random forests are ensembles of random trees, like bootstrapping with decision trees using randomly selected features.</p> Signup and view all the answers