Tidy Modeling

Question 1

Models are mathematical tools that can describe a system and capture

Answer 1

relationships in the data given to them

Question 2

Predicting future events, determining between-group differences, map-based visualizations, and pattern discovery are all

Answer 2

Purposes for which models can be used

Question 3

The utility of a model hinges on its ability to be

Answer 3

Reductive (reduce complex relationships to simpler terms)

Question 4

Purpose of a descriptive model

Answer 4

Describe or illustrate characteristics of some data

Question 5

Descriptive models need not have a purpose other than visually emphasizing an artifact in the data (T/F)

Answer 5

T

Question 6

Producing a decision for a research question or to explore a particular hypothesis is the goal of

Answer 6

Inferential models

Question 7

An inferential model starts with a predefined hypothesis about a population and produces a

Answer 7

Statistical conclusion (rejection of hypothesis, interval estimate, etc.)

Question 8

Inferential modeling techniques typically produce a __________ output

Answer 8

Probabilistic (p-value, CI, posterior probability)

Question 9

To compute probabilistic outputs, probabilistic assumptions must be made about the data and the underlying processes that generated the data because

Answer 9

The quality of statistical modeling is highly dependent on the pre-defined assumptions and how well the data fit them

Question 10

The primary goal of predictive models is that the predicted values have

Answer 10

The highest possible fidelity to the true value of the new data

Question 11

Problem type being resolved by predictive models is

Answer 11

Estimation

Question 12

In predictive models, more interest is vested in the predicted value than

Answer 12

Why the predicted value is what it is

Question 13

Predictive models can include measures of uncertainty (T/F)

Answer 13

True

Question 14

Most important factor affecting predictive models…

Answer 14

How the model was developed

Question 15

Predictive mechanistic models produce a model equation that

Answer 15

Depends on assumptions

Question 16

In predictive mechanistic models, data are used to estimate…

Answer 16

Unknown parameters of the model equation to generate predictions

Question 17

In predictive mechanistic models, differential equations are set based on

Answer 17

The model’s assumptions

Question 18

Unlike inferential models, predictive mechanistic models allow for data-driven statements on how well the model performs based on

Answer 18

How well it predicts the existing data

Question 19

Empirically-driven models are created with _____ assumptions

Answer 19

Vague

Question 20

Empirically-driven modeling most associated with _______ learning

Answer 20

Machine

Question 21

KNN modeling is an

Answer 21

Empirically-driven predictive model

Question 22

How does KNN work?

Answer 22

Given reference data, a new sample is predicted by using the values of K most similar data in the reference set

Question 23

In predictive models, if the structure of the model is good, then

Answer 23

The predictions would be close to the actual values

Question 24

Three types of models

Answer 24

Descriptive, inferential, and predictive

Question 25

Two types of predictive models

Answer 25

Mechanistic and empirically-driven

Question 26

Ordinary Linear Regression model is descriptive when

Answer 26

Restricted smoothing splines (similar to LOESS) are used to describe trends in data using OLR with specialized terms

Question 27

OLR is inferential when

Answer 27

Statistical results (p-values for ex) are used for inference

Question 28

OLR is predictive when

Answer 28

A simple linear regression produces accurate predictions

Question 29

KNN should not be used for inference because

Answer 29

Its nature makes the math required for inference impossible

Question 30

The predictive capacities of descriptive and inferential models should not be ignored because of how they model how

Answer 30

How variables relate to the probability of outcomes

Question 31

Predictive performance relates to how close the model’s

Answer 31

Fitted values are to the observed data

Question 32

Whether a model is appropriate cannot be determined by ______ alone

Answer 32

Statistical significance

Question 33

Unsupervised models learn patterns, clusters, or other characteristics of data (understand relationships between variables) but lack

Answer 33

An outcome (dependent variable)

Examples: principal component analysis (PCA), clustering, and autoencoders

Question 34

Supervised models have an outcome variable. Examples are…

Answer 34

Linear regression, neural networks, etc.

Question 35

Two sub-categories of supervised models

Answer 35

Regression (predictable numeric outcome)

Classification (predicts outcome based on ordered or unordered set of qualitative values)

Question 36

Outcomes (what is being predicted) are also known as…

Answer 36

Labels, endpoints, or dependent variables

Question 37

Independent variables (used to make predictions) also known as…

Answer 37

Predictors, features, or covariates

Question 38

Exploratory data analysis shows

Answer 38

How variables are related to each other (distributions, typical ranges, etc.)

Question 39

During EDA, two main questions should be answered, which are
…

Answer 39

How did I come by these data?

Are the data relevant to the problem?

Question 40

Performance metrics should be identified prior to

Answer 40

The analysis process

Question 41

Phases of Modeling

Answer 41

EDA (iterate between numerical analysis and visualization)

Feature engineering (use existent variables to create new variables)

Model tuning and selection (specifying or optimizing the structural parameters of models)

Model evaluation (assess Model performance, examine residual plots)

Question 42

A main effect is a Model term that contains a

Answer 42

Single predictor variable

Question 43

Root mean squared error (RMSE) is used in regression models by taking the difference (residuals) between the

Answer 43

Observed and predicted values in calculations

Question 44

Primary approach for empirical model validation is to split the existing pool of data into two distinct sets WHICH ARE

Answer 44

Training set - majority of data; used to build model
Test set - determines whether model is successful (should only be looked at once, or it becomes part of the modeling process)

Question 45

Simple random sampling is the most common method used to

Answer 45

Split data into training and test sets