Good afternoon everyone, we are group 7 and we would be reporting about
Trendlines and Regression Analysis.
Introduction
In today’s data-driven world, organizations and researchers rely on analytics to discover patterns,
predict trends, and support better decision-making. Among the fundamental tools in business
analytics, trendlines and regression analysis play a crucial role in identifying relationships between
variables and forecasting future outcomes. These techniques help managers and analysts
understand how one factor (known as the independent variable) affects another (the dependent
variable), allowing them to form evidence-based conclusions and strategic plans.
A trendline is a line plotted through a set of data points that reveals the general direction of change.
It visually represents whether the data show an upward, downward, or stable pattern over time.
Regression analysis, on the other hand, is a more advanced statistical method that estimates the
mathematical relationship between variables, often expressed through a linear equation of the
form:
Y =a+b X
or
Y^ = β0 + β 1 X
Here, β 0(beta zero) represents the intercept, β 1(beta one) represents the slope of the line, and X is
the independent variable used to predict the dependent variable Y . The goal of regression analysis
is to find the best-fitting line by estimating the values of β 0and β 1, minimizing the difference
between predicted and actual values. Here is the formula to get the coefficients:
_____________________________________________________________________________________
MOTIVATING EXAMPLES
To better understand the usefulness of trendlines and regression analysis, the following examples
demonstrate how these techniques can be applied to solve real-world problems.
1. Adipose Tissue Problem
Researchers want to find out whether waist circumference (WC) can be used to predict abdominal
adipose tissue (AT) (measured in cm²).
Because CT scans are costly, they want a simple formula to estimate AT using waist measurements.
This is the formula for regression analysis, so we have to see our formula for regression.
Ȳ^ = β0 + β 1 X
X → Waist Circumference (cm)
Y → Adipose Tissue (cm²)
X → Waist Circumference
Ȳ → Mean of Adipose Tissue
β₁ → Slope (change in Y for each 1 cm change in X)
β₀ → Intercept (predicted Y when X = 0)
Since the goal of regression analysis is to find the values of β 0and β 1 here is what we will use
The first step is to compute the means so:
Next would be to make a deviation, I suggest making a table to organize and get the values easier
Now we can compute for the β 1 or the slope
and so we have our β₁ which is 3.
Now the next step is to compute the β₀ or the intercept
And there we have it β₀ = −105.
Now we can make our Regression Equation which is:
Y^ =−105+3 X
2. Newspaper Problem
A publishing company owned by Mr. Warner Sr. wants to estimate the Sunday circulation of its new
weekend edition that was proposed by his son. They know their daily newspaper sells about 500,000
copies, but they want to predict how many readers they might have on Sunday.
To do this, they collected data from 4 other newspapers that publish both daily and Sunday
editions.
X → Daily Newspaper Circulation (in thousands)
Y → Sunday Newspaper Circulation (in thousands)
X̄ → Mean of Daily Circulation
Ȳ → Mean of Sunday Circulation
β₁ → Slope (change in Y for each 1 unit change in X)
β₀ → Intercept (predicted Y when X = 0)
The first step is to compute the means so:
X̄ = 475, Ȳ = 670
Next would be to make a deviation
Now we can compute for the β 1 or the slope
and so we have our β₁ which is 0.96
Now the next step is to compute the β₀ or the intercept
And there we have it β₀ = 214.
Now we can make our Regression Equation which is:
Y^ =214+0.96 X So ,the if the daily newspaper sells 500,000 copies, the predicted Sunday circulation
is:
Y^ =214+0.96 (500)=214+ 480=694 or 694,000 copies
METHODS OF REGRESSION
What is Regression?
Regression is a statistical technique used to describe and analyze the relationship between one dependent
variable and one or more independent variables. It allows researchers to explain how changes in the
predictors (independent variables) affect the outcome (dependent variable).
In its simplest form, simple linear regression involves one predictor variable and estimates the best-
fitting straight line through the data.
Linear Regression Model
Y = β0 + β 1 X 1 + β 2 X 2 +...+ β k X k + ε