Get Demand Analytics Coursera Quiz Answers
Demand Analytics Coursera Quiz Answers
Week 1: General principles
Q1. Poor demand planning and forecasting can result in
- Excessive inventory
- Over-time production
- Backorder or lost sales (revenue)
- Expedited shipping
Q2. It can be hard to predict demand accurately because demand can be influenced by many factors, such as
- Price and promotion
Q3. How do the views on demand planning and forecasting differ between operations and marketing?
- Marketing likes to exaggerate demand but operations tends to be more conservation in demand estimation
- Both marketing and operations like to exaggerate demand and strive to meet all demand without considering the cost
- Both marketing and operations like to be conservative in demand estimation in order to be cost efficient.
Q4. Demand analytics is the art and science of applying analytics to demand planning and forecasting; it can
- better match demand with supply.
- increase revenue and reduce cost.
- improve customer satisfaction and reduce production, inventory and shipping costs.
Week 2: Predicting trend
Q1. The least squares methodology in model estimation means
- minimizing the sum of squares of all errors (an error is the difference between an observed value and a predicted value).
- minimizing the sum of all errors.
- minimizing the sum of all absolute errors.
Q2. What is the meaning of R square = 0.439?
- 56.1% of the variation in the data comes from random errors that cannot be explained by the model.
- The model can explain 43.9% of the variation in the data.
Q3. The p-value of the slope indicates whether the regression model works better than the arithmetic mean
- the smaller the p-value is, the better a predictor the regression equation is than the arithmetic mean.
- The larger the p-value is, the better a predictor the regression equation is than the arithmetic mean.
- Neither is true.
Q4. The most popular patterns in demand are
- Trend over time, either linear or nonlinear.
- Fluctuations that combine trend and seasonality.
- Seasonal pattern with periodical peaks and valleys.
Week 3: Demand Analytics
Quiz 1: Model validation and improvement
Q1. What is a residual in a regression model?
- A residual is the difference between an actual observation and the predicted value.
- Residuals are the errors that are not explained by the model.
Q2. What are the causes of residuals?
- Errors in the model: the model misses some variables.
- Unpredictable random noises which cannot be eliminated.
- Errors in the model: The model has the wrong form, for instance, the model should be nonlinear rather than linear.
Q3. The linearity condition is met if
- The residuals look like white noises.
- The residuals are purely randomly distributed.
- The residuals have a trend or a cyclic pattern.
Q4. The independence condition is met if
- The residuals are not correlated over any of the observations, meaning, you cannot predict the next residual based on existing residuals.
- The residuals are independent over all observations.
Q5. The equal variance condition is met if
- The residuals have constant variance over all observations.
- The variance of the residuals may change over observations.
Quiz 2: Multiple regression
Q1. The Coefficients Table of the model shown in the videos of this lesson is given as follows:
Which variable(s) are significant at a 10% significance level?
- Total home sales (millions)
- Time in month
- 10-Piece Set sale price
Q2. Given the same Coefficient Table as in Question 1, what is the meaning (in words) of the coefficient, 12.48, for Time in month?
- On average, the sold units increase by 12.48 units for each month.
- On average, the sales in $ increase by $12.48 for each month.
- On average, the sold units decrease by 12.48 units for each month.
Q3. Given the same Coefficient Table as in Question 1, what is the meaning (in words) of the coefficient, -6.27 (using only 2-digit), for 10-Piece Set sale price?
- On average, the sold units decrease by 6.27 units for every $ of price drop.
- On average, the sold units increase by 6.27 units for every $ of price drop.
- On average, the sales in $ increase by $6.27 for every $ of price drop.
Week 4: Modeling and formatting categorical variables
Q1. When to use categorical variables?
- The variable can only choose discrete values, such as 1 or 0 (yes or no)
- The variable can choose any fraction
Q2. How to model a categorical variable?
- For each option (or category) of the variable, we should set up a binary variable
- We select some but not all options (or categories) of the variable, and set up a binary variable for each option (or category).
- We can set up a variable with multiple integer values each corresponding to an option (or category).