Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Unit 6: Time-based Data

Time Series

A Time Series is a sequence of numerical data points in successive order. Usually, a time series is a sequence taken at successive equally spaced points in time. Time Series Analysis involves methods for analyzing time series data to extract meaningful statistics and characteristics of the data.

Components of Time Series

A time series may consist of the following four components:

  1. Secular Trend (T): The long-term tendency of the data to increase or decrease. For example, population growth over decades.
  2. Seasonal Variation (S): Short-term fluctuations that occur regularly during a year, month, or week. E.g., increased sales of winter clothes in December.
  3. Cyclical Variation (C): Oscillating movements lasting more than a year. These are typical of business cycles (prosperity, recession, depression, recovery).
  4. Irregular Variation (I): Unpredictable, random variations due to unforeseen events like strikes, natural disasters, or wars.
Time Value Trend Actual Data

Methods of Measuring Trend

To isolate and measure the secular trend component, we typically use the following methods:

1. Moving Average Method The moving average is created by averaging the data points over a specified period (e.g., 3-yearly or 5-yearly moving average). It smooths out short-term fluctuations.

2. Method of Least Squares This method provides a mathematical equation to fit the trend exactly. For a straight-line trend \( y = a + bx \), we find the constants \( a \) and \( b \) by solving the normal equations:

\[ \sum y = n a + b \sum x \] \[ \sum xy = a \sum x + b \sum x^2 \]

Competency-Based Questions

Q1. A firm observing its 5-year sales data notices an unexpected dip in sales in 2020 due to the pandemic lockdowns. Identify which component of the Time Series this dip represents. Does it represent cyclical variation? Justify your answer.

Q2. The profit (in lakhs) of a retail company from 2018 to 2022 is 12, 16, 22, 28, and 30, respectively. Find the 3-yearly moving averages to calculate the trend values.

Q3. For the dataset in Q2, establish the equation for a straight-line trend using the Method of Least Squares. Estimate the profit for the year 2025 using the trend line equation.

Q4. Why is the Method of Least Squares considered mathematically more rigorous and reliable for forecasting than the Moving Average Method? Discuss the limitations of the Moving Average Method.

Answers

Ans 1. Irregular Variation. The dip is due to an unforeseen, unpredictable event (the pandemic), unlike Cyclical Variation which has natural rhythmic economic cycles lasting more than a year.

Ans 2. Moving averages: For 2019: \( \frac{12+16+22}{3} = 16.67 \). For 2020: \( \frac{16+22+28}{3} = 22 \). For 2021: \( \frac{22+28+30}{3} = 26.67 \).

Ans 3. Taking mid-year 2020 as Origin (X = 0), \( x \) values are -2, -1, 0, 1, 2. \( \sum x = 0 \), \( \sum y = 108 \), \( \sum x^2 = 10 \), \( \sum xy = 46 \). \( a = \frac{108}{5} = 21.6 \), \( b = \frac{46}{10} = 4.6 \). Line: \( y = 21.6 + 4.6x \). For 2025 (\( x = 5 \)): \( y = 21.6 + 4.6(5) = 44.6 \) lakhs.

Ans 4. Method of Least Squares uses a strict mathematical formula giving an exact trend line free from subjective bias, and it can forecast future values. Moving Average cannot estimate extreme (start/end) boundary values and ignores the exact mathematical curve best fit.