Practical Statistics for Data Scientists (part 2)

13 flashcards · Computer Science · by charlie42

  1. The number of years a worker was exposed to cotton dust (Exposure) versus a measure of lung capacity (PEFR or “peak expiratory flow rate”). Linear regression model: How can we interpret the intercept (b0) and the regression coefficient (b1)?

    The intercept, or b0, is 424.583 and can be interpreted as the predicted PEFR for a worker with zero years exposure. The regression coefficient, or b1, can be interpreted as follows: for each additional year that a worker is exposed to cotton dust, the worker’s PEFR measurement is reduced by –4.185.

  2. What is the “hat” notation is used for?

    To differentiate between estimates and known values

  3. Formula to compute a residual for linear regression?

  4. What does this formula calculate?

    Residual sum of squares or RSS

  5. The method of minimizing the sum of the squared residuals for estimating the linear regression model is termed (2)

    Least squares regression, or ordinary least squares (OLS) regression

  6. Suppose you're predicting house prices. By how much the price will increase, if the square footage is increased by 1000?

    1000 * 2.288e+02 = 1000 * 228.8 = 228,800

  7. Most common linear regression performance metric and it's formula

    Root mean squared error

  8. Another name for R-squared statistic

    Coefficient of determination

  9. What does R-squared statistic measure (2)?

    The proportion of variation in the data that is accounted for in the model, or how well the model fits the data

  10. Formula for R-squared statistic

  11. Formula for t-statistic in linear regression and what it's used for

    It's used for variable selection (higher t-statistic - more useful the variable)

  12. Visualize k-fold cross-validation algorithm (k=5)

  13. In a linear regression model, what is b1 called?

    Regression coefficient