MBA/MBA Semester 1/Statistics for Management/Module 5: Forecasting Techniques

Module 5: Forecasting Techniques โ€” Quiz Bank

Lesson 3: Covariance and Correlation Overview

QUESTION 1 OF 81
Covariance indicates:
โœ… Direction of relationship
Covariance tells whether variables move together (positive) or oppositely (negative); correlation gives the degree.
QUESTION 2 OF 81
Correlation measures:
โœ… Both strength and direction of a linear relationship
The coefficient's size shows strength and its sign shows direction, for linear relationships.
QUESTION 3 OF 81
Positive covariance means:
โœ… Both variables move in the same direction
Positive ฮฃ(x - xฬ„)(y - ศณ) means the variables rise and fall together.
QUESTION 4 OF 81
Covariance is limited when comparing across datasets because of:
โœ… Scale dependence
Covariance's size depends on the units of the variables, so correlation is used to compare.
QUESTION 5 OF 81
Pearson correlation coefficient ranges from:
โœ… -1 to 1
-1 is perfect negative and +1 perfect positive correlation, with 0 meaning none.
QUESTION 6 OF 81
A Pearson R value of -1 indicates:
โœ… Perfect negative linear correlation
-1 means the variables move exactly oppositely along a straight line.
QUESTION 7 OF 81
In finance, correlation helps in:
โœ… Asset diversification
Choosing assets with low or negative correlation diversifies a portfolio and reduces risk.
QUESTION 8 OF 81
Covariance is primarily used for:
โœ… Portfolio risk analysis
Covariances between asset returns feed directly into portfolio variance (risk).
QUESTION 9 OF 81
The main types of correlation are:
โœ… Simple, multiple, partial
The lesson classes correlation as simple, multiple and partial (or total), alongside linear/nonlinear and positive/negative.
QUESTION 10 OF 81
Partial correlation controls for:
โœ… Confounding variables
Partial correlation measures the relation between two variables while holding other influencing variables constant.
QUESTION 11 OF 81
Karl Pearson's coefficient is used for:
โœ… Measuring linear correlation
Pearson's r = Cov(X, Y) / (ฯƒX ฯƒY) measures the strength and direction of a linear relationship.
QUESTION 12 OF 81
The coefficient is unaffected by:
โœ… Measurement units
r is a dimensionless number, so changing the units (or origin and scale) does not change it.
QUESTION 13 OF 81
Spearman's rank correlation is used when:
โœ… Data is ordinal or non-linear
Spearman works on ranks, so it suits ordinal data and monotonic or nonlinear relationships.
QUESTION 14 OF 81
Spearman's coefficient uses:
โœ… Ranked data
It is computed from the ranks of the observations (rank differences d).
QUESTION 15 OF 81
Pearson is ideal for:
โœ… Linear relationships
Pearson measures the linear link between two continuous variables.
QUESTION 16 OF 81
Spearman is more robust when:
โœ… Data has outliers
Ranks limit the influence of extreme values, so Spearman is frequently used for data with outliers.
QUESTION 17 OF 81
Regression analysis predicts:
โœ… One variable based on another
Regression builds a model to predict the dependent variable from the independent variable.
QUESTION 18 OF 81
In simple linear regression, the model is:
โœ… Y = a + bX
The lesson writes the fitted line as Y = a + bX (a = intercept, b = slope).
QUESTION 19 OF 81
Least square method minimizes:
โœ… Sum of squared residuals
It picks the line whose total squared vertical deviations from the data are smallest.
QUESTION 20 OF 81
It fits the line of:
โœ… Best fit
The least-squares line is the "best fit" line.
QUESTION 21 OF 81
Curve fitting finds:
โœ… Smooth function approximating data
Curve fitting finds a smooth function (curve) that approximates the observed data.
QUESTION 22 OF 81
Polynomial fitting is used for:
โœ… Non-linear trends
Polynomial terms (t, tยฒ, ...) capture curved, non-linear trends.
QUESTION 23 OF 81
Rยฒ (coefficient of determination) shows:
โœ… Fraction of variance explained
Rยฒ is the share of total variation in Y that the model explains.
QUESTION 24 OF 81
A good model has Rยฒ close to:
โœ… 1
Rยฒ near 1 means almost all variation is explained.
QUESTION 25 OF 81
Standard error of estimate measures:
โœ… Residual variability
It measures how widely observed values scatter around the regression line.
QUESTION 26 OF 81
Lower SE indicates:
โœ… Stronger fit
Smaller scatter around the line means predictions are closer to observations.
QUESTION 27 OF 81
Coefficient of determination is denoted as:
โœ… Rยฒ
It is written Rยฒ or rยฒ ("R squared").
QUESTION 28 OF 81
It ranges from:
โœ… 0 to 1
In the usual cases the coefficient of determination lies between 0 and 1.
QUESTION 29 OF 81
Time series data is ordered:
โœ… Chronologically
A time series is observations taken at periodic points in time, in time order.
QUESTION 30 OF 81
Time series analysis is used for:
โœ… Predicting future based on historical trends
It studies past behaviour (trend, seasonal, cyclic) to forecast the future.
QUESTION 31 OF 81
Secular trend reflects:
โœ… Long-term movement in time series
A secular trend is the general rise, fall or constancy of data over an extended period.
QUESTION 32 OF 81
Trend line helps in:
โœ… Smoothing long-term changes
A trend line summarises the long-term direction, smoothing out short-term fluctuations.
QUESTION 33 OF 81
Seasonal variation recurs:
โœ… At regular intervals within a year
Seasonal variations are short-period cycles (under a year) that repeat regularly, caused by climate and customs.
QUESTION 34 OF 81
Seasonal effect is best analysed using:
โœ… Ratio-to-trend method
The ratio-to-trend method isolates seasonal indices relative to the trend.
QUESTION 35 OF 81
Cyclical variations occur due to:
โœ… Business cycles
Cyclical variations follow trade (business) cycles of boom, recession, depression and recovery lasting more than a year.
QUESTION 36 OF 81
Irregular variations are:
โœ… Unsystematic and random
They are unpredictable, random movements not explained by trend, seasonal or cyclical components.
QUESTION 37 OF 81
Numerical trend methods include:
โœ… All of the above
Curve fitting, moving averages and percentage change are all numerical trend methods.
QUESTION 38 OF 81
Moving average helps:
โœ… All of the above
It removes seasonality, supports short-term trend forecasting and smooths noise.
QUESTION 39 OF 81
Cyclical variation is measured by:
โœ… Residual component after trend and seasonal removal
The residual method removes trend and seasonal parts (and smooths the irregular) to leave the cyclical component.
QUESTION 40 OF 81
Irregular variation is usually:
โœ… Random and difficult to predict
It is the error term caused by natural forces and cannot be forecast.
QUESTION 41 OF 81
Retailers use trend analysis to:
โœ… Optimize pricing and inventory planning
Spotting sales trends lets retailers plan prices and stock levels.
QUESTION 42 OF 81
Trend analysis helps identify:
โœ… Long-term consumer behavior
It reveals the direction of data over time, such as long-term buying behaviour.
QUESTION 43 OF 81
A logistics firm used time series to:
โœ… Predict seasonal shipment volumes
Time series reveals seasonal patterns, so shipment volumes can be forecast.
QUESTION 44 OF 81
In a bank, time series helped:
โœ… Understand daily cash flow trends
Daily cash flow data over time shows trends useful for planning liquidity.
QUESTION 45 OF 81
What is the definition of correlation according to Croxton and Cowden?
โœ… Correlation is a statistical tool for discovering and measuring the relationship between two quantitative variables
Croxton and Cowden define correlation as the tool for discovering and measuring a quantitative relationship.
QUESTION 46 OF 81
Which statement about correlation is true based on the example provided?
โœ… Correlation measures the extent to which two or more variables fluctuate in relation to one another
Correlation studies how much variables fluctuate together, not causation.
QUESTION 47 OF 81
In which scenario would Spearman rank correlation be most appropriate?
โœ… When the data is provided in the form of rankings for various criteria
Spearman measures agreement between rankings.
QUESTION 48 OF 81
What is the primary goal of regression analysis?
โœ… To create a mathematical model illustrating the relationship between variables
Regression models the relationship so one variable can be predicted from another.
QUESTION 49 OF 81
What is the primary role of errors in regression analysis?
โœ… Errors consist of the random component and factors not considered in the statistical model
Errors are the random and unexplained part left after the systematic component.
QUESTION 50 OF 81
What is a cyclical variation in time series analysis?
โœ… Long-term pattern in data over multiple years
Cyclical variation spans more than a year, typically three to ten years.
QUESTION 51 OF 81
Which of the following is a stage of a business cycle?
โœ… Expansion
Expansion (boom) is the platform's answer; the lesson names boom, recession, depression, recovery.
QUESTION 52 OF 81
R-squared represents:
โœ… The percentage of variation in the dependent variable explained by the independent variable(s)
Rยฒ is the share of variation explained by the model.
QUESTION 53 OF 81
When can Rยฒ take negative values?
โœ… When there is no intercept in linear regression
Regression without an intercept (or poor non-fitted predictions) can give negative Rยฒ.
QUESTION 54 OF 81
The main causes of seasonal variations in time series include:
โœ… Climatic conditions and customs
Seasonal variation arises from climate and customs/traditions.
QUESTION 55 OF 81
In time series forecasting, a horizontal trend usually indicates:
โœ… No significant change
A horizontal trend is flat: no significant change.
QUESTION 56 OF 81
The probability of committing a Type I error is represented by:
โœ… Alpha (ฮฑ)
Type I error probability is ฮฑ.
QUESTION 57 OF 81
What does a Type II error indicate?
โœ… Accepting a false null hypothesis
Type II error (ฮฒ) is failing to reject a false null hypothesis.
QUESTION 58 OF 81
The period of time between two consecutive phases of a business cycle is known as:
โœ… Duration of cyclical variations
The lesson calls it the period (duration) of cyclical variations.
QUESTION 59 OF 81
If a series shows both seasonal and cyclical fluctuations, the time between successive measurements is usually:
โœ… Less than or equal to a year
The lesson says seasonal fluctuations appear when the interval is one year or less.
QUESTION 60 OF 81
The standard error decreases when:
โœ… The sample size increases
Standard error = ฯƒ/โˆšn, so it falls as n rises.
QUESTION 61 OF 81
A confidence interval of 95% means:
โœ… There is a 95% chance that the true parameter lies within the interval
The platform's interpretation of a 95% confidence level.
QUESTION 62 OF 81
In time series analysis, secular trends are associated with:
โœ… Long-term consistent patterns
A secular trend is the long-term tendency to rise, fall or stay constant.
QUESTION 63 OF 81
Which can influence the width of a confidence interval?
โœ… All of the above
Sample size, significance level and sample variability all affect width.
QUESTION 64 OF 81
The method used to examine the general tendency of data to rise or fall over an extended period is called:
โœ… Secular Trends Examination
That is the secular trend (Lesson 19).
QUESTION 65 OF 81
What is the meaning of a positive value of the correlation coefficient r?
โœ… the variables x and y increases or decrease together
Positive r means the variables move together.
QUESTION 66 OF 81
What is regression analysis used for?
โœ… to deal with the formulation of mathematical model depicting relationship amongst variables
Regression builds a mathematical model of the relationship.
QUESTION 67 OF 81
What are dx and dy in the Assumed Mean Method formula?
โœ… Deviations of X from its assumed mean and deviations of Y from its assumed mean
dx and dy are deviations from the assumed means.
QUESTION 68 OF 81
What does the symbol Y-hat represent in the regression model?
โœ… Predicted value of Y
ลถ is the predicted (fitted) value of Y.
QUESTION 69 OF 81
How are the constants a and b determined?
โœ… by using the normal equations
Least squares normal equations give a and b.
QUESTION 70 OF 81
What does no correlation indicate in a scatter diagram?
โœ… The dots plotted on a graph do not follow a particular direction
No pattern means no correlation.
QUESTION 71 OF 81
What does r = -1 imply?
โœ… all data points lie on a line for which Y decreases as X increases
Perfect negative linear correlation.
QUESTION 72 OF 81
What are the two kinds of correlation based on direction of change?
โœ… Positive Correlation and Negative Correlation
Direction gives positive and negative correlation.
QUESTION 73 OF 81
What is the range of the correlation coefficient?
โœ… The value lies between +1 and -1
-1 <= r <= +1.
QUESTION 74 OF 81
What is the purpose of the method of least squares?
โœ… fitting a linear equation to the given data
It fits the best straight line by minimising squared errors.
QUESTION 75 OF 81
What is multiple regression?
โœ… When at least three variables are studied and their relationship is simultaneously worked out
Multiple regression involves three or more variables simultaneously.
QUESTION 76 OF 81
In y = 2x + 3, what does the coefficient 2 represent?
โœ… The slope of the regression line
The coefficient of x is the slope.
QUESTION 77 OF 81
Which scatter plot indicates the strongest positive correlation?
โœ… Points forming a straight line sloping upward from left to right
A rising straight line means r close to +1.
QUESTION 78 OF 81
In y = 2x + 3, what is the predicted y when x = 5?
โœ… 13
2(5) + 3 = 13.
QUESTION 79 OF 81
In Y = 75.65 + 0.50X, the intercept is:
โœ… 75.65
The constant term is the intercept.
QUESTION 80 OF 81
In Y = 82.51 + 0.50X, the slope is:
โœ… 0.5
The coefficient of X is the slope.
QUESTION 81 OF 81
Coefficient of correlation if the two regression coefficients are 0.5 and 4.5:
โœ… 1.5
r = sqrt(bxy x byx) = sqrt(2.25) = 1.5 (platform key as answered; note r cannot exceed 1, so the data is flawed).