ISYE 6501 EXAM Study Guide Questions And Answers Verified 100% Correct

ISYE 6501 EXAM Study Guide Questions And Answers Verified 100% Correct what is a common % of data split for training vs. validation? what about training vs. validation vs. testing? - ANSWER around 70/30 common for training/validation. 70/15/15 or somewhere around that number is common for training vs validation vs testing if someone describes a time series model as 'double exponential smoothing', which components will be involved? which one will NOT be involved - ANSWER level and trend will be involved. seasonality would not be how are ARIMA and GARCH similar? how are they different? - ANSWER ARIMA and GARCH both can be used for time series data and smoothing. assuming ARIMA(p, d, q), what values for p, d, and q would essentially represent a basic exponential smoothing model - ANSWER ARIMA (0, 1, 1) Reference = lecture slides When does ARIMA perform better than exponential smoothing in terms of short-term forecasting? how many samples are generally required for ARIMA model? - ANSWER when the data is more stable, with fewer peaks, valleys, and outliers. want 40 PAST data points for ARIMA to work well. which set of the following equation is used to detect a decrease vs. increase in the context of CUSUM analysis St = max{0, St-1 + (Xt - u - C)} Is St = T vs St = max(0, St-1 + (u - Xt - C)} Is St = T? - ANSWER Detecting an increase St = max{0, St-1 + (Xt - u - C)}Is St = T Detecting a decrease St = max(0, St-1 + (u - Xt - C)} Is St = T See ISYE-6501 Module 6(L1-L2) Playlist v0822 timestamp 5:05 / 5:49 in a regression model, given two variables A, B, what does it mean if the coefficient is a lot larger (and positive) for variable A (years of education), relative to variable B (age in years) e.g. income = a0 + a1 (years of education) + a2 (age in years) - ANSWER larger coefficient (assuming coefficient for both variables are positive) would represent bigger impact on the predictor variable two lm models. A has R^2 of 0.99 whereas B has R^2 of 0.80 on the training data set, which one will likely have better performance on the validation or test data set? - ANSWER it is not possible to tell What model would you use for any nonseasonal series of numbers that exhibits patterns and is not a series of random events? a. ARIMA g. k-means clustering b. CART h. k-nearest-neighbor classification c. Crossvalidation i. Linear regression d. CUSUM j. Logistic regression e. Exponential smoothing k. Support vector machine f. GARCH - ANSWER a If you wanted to forecast volatility of stock data, which model would you choose? Keep in mind, this data may have high heteroskedasticity. a. ARIMA g. k-means clustering b. CART h. k-nearest-neighbor classification c. Crossvalidation i. Linear regression d. CUSUM j. Logistic regression e. Exponential smoothing k. Support vector machine f. GARCH - ANSWER f Which model is a decision tree where each fork is a split in a predictor variable and each end node contains a prediction for the outcome variable? a. ARIMA g. k-means clustering b. CART h. k-nearest-neighbor classification c. Crossvalidation i. Linear regression d. CUSUM j. Logistic regression e. Exponential smoothing k. Support vector machine f. GARCH - ANSWER b What is the difference between logistical and linear regression models? - ANSWER Linear regression is used for predicting the continuous dependent variable using a given set of independent features whereas Logistic Regression is used to predict the categorical. Linear regression is used to solve regression problems whereas logistic regression is used to solve classification problems. What model is often used when differentiating between two things? Often the ANSWER is "yes" or "no" - ANSWER Classification Numbers with a meaning, ex.) age would be known as what type of data? - ANSWER Quantitative data Data that can only take one of two values is known as what type of data? - ANSWER Binary data Data where there is no relationship between data points, ex.) different loan applicants is known as? - ANSWER Unrelated data Same data recorded over time, often recorded at equal intervals, ex.) daily sales, stock prices, child's height on each birthday is known as? - ANSWER Time series data In time series modeling, how good values of alpha, beta and gamma identified? - ANSWER Model is optimized by minimizing (Ft-xt)^2 You have just trained a model and selected the best one using 10-fold cross validation. For your final model, you should re-train the model using: a. The average of all of the training data in each fold b. The training data & validation data from each fold c. The entire training & validation set d. The entire dataset - ANSWER c. The entire training & validation set. The test set should still be reserved to measure final model performance. In a random forest model, each tree will utilize the same subset of factors of size m = 1 + ln(n), where n = total number of factors: True False - ANSWER False. Each tree will utilize a random set of factors of size m = 1 + ln(n). You might use this method of adjusting data for data in a bounded range. Ex.) neural networks, batting averages, SAT scores - ANSWER Scaling What are the two types of patterns data has? - ANSWER Real effect and random effect You can't judge a model based on how well it fits the training data because of what reason? - ANSWER Random effects Support Vector Machine(SVM) is a supervised machine learning algorithm used for? -