Regression

Question 1

KNN Classification - training and predicting

Answer 1

Training: store all the data
Prediction:
1. calculate the distance from x to all points in your dataset
2. sort the points in your dataset by increasing distance from x
3. predict the majority label of the k closest point

Distance
- euclidean distance, manhattan distance, cosine distance = 1 - cosine similarity

Question 2

KNN Regression

Answer 2

We take the average of the k nearest items

Question 3

KNN Hyperparameters

Answer 3

K

- how many amount of neighbors we’re using. General rule: k=sqrt(n). Then do your grid search from here.

Question 4

KNN - noise vs signal

Answer 4

lower k tends to overfit

higher k tends to underfit - capture less noise, and low signal

Question 5

standardization

Answer 5

use standardization when your data has varying scales
(data point - mean) / standard deviation

put each scale to mean = 0

Question 6

KNN (pros, cons)

Answer 6

pros:
- super simple
- training is trivial
- easy o add more data
- few hyperparameter

cons

• if you have a lot of features, you need a lot more data, but can be costly to gather more data
• high prediction cost
• bad with high dimensions. Anything more than 5 is bad.
• categorical features don’t work well

Question 7

Mean Squared Error (MSE)

Answer 7

expected value of the square of the error
MSE = 1/n * E( predicted - actual)**2
- the average squared difference between the estimated values and the actual values
- mean of all the square errors in your model and data

Question 8

Irreducible error

Answer 8

Error that we can’t do anything about Even if we had all possible data and could build a perfect model, we can’t predict values exactly.

Question 9

Bias and variance

Answer 9

• errors that we can control

bias = failing to capture some of the signal (underfit)
variance = error we get when from real world data. Where are the errors coming from and How consistently we're off.

When capturing more signal, you’re naturally capture more noise and variance.

Question 10

k fold cross validation

Answer 10

train test split - reserve data for the ultimate testing set.
then do k-fold training: training and validation set

Question 11

churn

Answer 11

decision rules

Question 12

Linear Regression - scatterplot

Answer 12

good practice to plot a scatter plot, if it appears a linear relationship, between dependent and independent variables, it is good hint that we can use linear regression as a learning algorithm.

Question 13

Feature Engineering

Answer 13

anytime you use your current features to create new features.

Question 14

Linear Regression with single feature

Answer 14

y = mx + b

Question 15

Linear Regression - how to pick the best line

Answer 15

Residual = the distance between our predicted value and the actual value

Find the line that minimizes the total sum of squared residuals.

Question 16

Linear Regression with multiple features

Answer 16

the linear relationship is:

y = B01 + B1x1 + B2x2 + Bpxp

Question 17

Linear Regression - R-squared

Answer 17

give you a sense of how well your model performs - how much variance of the data you’re capturing. It tells you how much better your model is doing better than the dumb model. Close to 0 = dumb model. 1 = overfitting

Something to watch out for: the more features you have, the higher r-squared is. Hence, people adjusted r-squared is a better measure

Question 18

Linear Regression - Adjusted r-squared

Answer 18

normalized to the increase of features. It is a better measure of accuracy of your model because r-squared is susceptible to number of features

Question 19

How do you know if a feature and dependent variable has a linear relationship?

Answer 19

You can plot the residuals (y_predicted - y_actual). We want to see our residuals normally distributed around 0 and shows no trend.

Question 20

studentized residuals

Answer 20

divide residual by the estimate of standard deviation of the residuals.
It helps you to find outliers. You can remove a point from the data, plot the residual plot again to find outliers.

Question 21

homoscedasticity vs. heteroscedasticity

Answer 21

homoscedasticity = When variance of your residual is constant.

how to test?

1. Divide your data into 2 parts, if variance on left equals to right, it’s homo, else, it’s heteroscedasiticity
2. sm.stats.diagnostic.het_goldfeldquandt then look for p-value. Null hypothesis = homoscedasticity

Question 22

Normality

Answer 22

residuals are normally distributed

To see if residuals are normally distributed, we look at QQ plot - If they follow the same line, it is normal.

We can also use Jarque-Bera(JB). Look at p-value.
Null: it follows normality
Alt: doesn’t follow normality

Question 23

Multicollinearity

Answer 23

When you have two features or more that depend on each other. Ie, height and weight.

• It means you can’t be confidence about your features so you will have to remove one of more features to be confident about your feature
• Variance inflation factor VIF. If VIF > 10, it’s a good indication that you have multiplelinearity

Question 24

Linear Regression (prediction vs inferential)

Answer 24

1. when there’s a linear relation between outcome and features
2. Goal: find the coefficients of the features
3. what coefficients should I choose to minimize my sum of squared residuals? - the difference between my observations and predictions

When I’m making prediction, and I don’t care about inferring the coefficients. I build a linear regression model and predict.

If you want to interpret the data - if you want to see how confidence you are with each coefficient, then you have to follow the following conditions.

1. Linearity
2. Homoscedascity
3. Multicolinearity
4. Normality
5. All data is i.i.d (individually independent distributed)

Question 25

Linear Regression - Top bottom, bottom top

Answer 25

bottom top: start by each feature, start with the one giving you the lowest RMSE, then add second, add third and see if you can reduce your RMSE. top bottom: consider all features, remove one by one

Question 26

Regularized regression - what does it do?

Answer 26

It introduces parameters into linear regression to allow us to adjust for biases and variances tradeoff. It introduce an alpha to the linear regression equation. The bigger the alpha, the smaller beta gets. When beta is small, the model is less sensitive to each individual feature, it is less likely to overfit.

Question 27

regularized regression - criteria

Answer 27

All predictors need to be on the same scale. | standardized_feature = (raw_feature - mean(raw_feature))/st_dev ( raw_feature)

Question 28

ridge regression

Answer 28

optimization: alpha(beta**2) | As alpha increases, it pushes all betas proportionately close to 0

Question 29

overfitting and underfitting

Answer 29

When overfitting, we're capturing a lot of signal but we're also introducing a lot of noise. So when we fit our real world data into our model, it is going to give me inconsistent outcome.

Question 30

lasso regression

Answer 30

optimization: alpha( |beta| ) As alpha increase, some betas drop to 0 sooner than other betas. Benefit: allow for feature selection - I believe there are features that matter more than others, and I want the features that matter more to stay in the model longer as I'm increasing alpha, and features that matter less to drop out.

Question 31

Logistic Regression - decision boundary

Answer 31

setting a threshold to determine class 1 and class 2. We set this boundary based on business needs

Question 32

Logistic Regression - find the best model

Answer 32

minimize log loss

Question 33

how do you define which model is better in logistic regression?

Answer 33

ROC curve it works like this: false positive rate on x, true positive rate on y. Then start moving threshold to see how the fpr and tpr rate changes. The bigger the area of this curve, the better we're doing with our model.

Question 34

logistic regression - performance

Answer 34

depends on what class you're interested in, we need to adjust our performance measurement using a confusion matrix. For example, if positive is more important, we want to reduce false negative, etc. Whenever accuracy is not good indicator, always use confusion matrix and other measurement like recall, precision, etc.

Question 35

linear regression - performance

Answer 35

whichever model with the lowest RMSE is the better model

Question 36

what does it mean that logistic regression is a soft classifier?

Answer 36

able to provide probabilities of an outcome belong to one class vs the other.

Question 37

confusion matrix

Answer 37

it is a table used to describe the performance of a classification model. Accuracy = how often is the classifier correct? Recall = when it's actually yes, how often does it predict yes? Precision = when it predicts yes, how often is it correct?

Question 38

how do you deal with imbalanced class?

Answer 38

Before you do any of these, make sure to do train test split. Real world data should reflect real data as accurately as possible. 1. oversampling - one way to do it is to bootstrap 2. undersampling - appropriate when you have more data than you can process. Or when data in the majority class is less important to the result. ie. 1 million rows, 100,000 in one class. We randomly make 100,000 data from the majority class. 3. SMOTE - it generates some new data points looking at two samples of data but usually results is not good.

Question 39

using a confusion matrix in a business case

Answer 39

1. build a model and create a confusion matrix 2. assign dollar value to a confusion matrix with a business use case 3. multiply them together to calculate profit 4. test it with a range of threshold to see which threshold gives us the highest revenue