Report of

Forecasting Rice Production

With Regression Analysis
Submitted To
… JINNAH UNIVERSITY FOR
WOMEN…

Course Instructor
MISS HUMA ABU BAKAR

… Specialization: Managerial
economics…
Course Code: ECO 3071
Submitted By
Muniba Khan

1
BS 3rd Year Commerce

2
June 11, 2024

3
Regression analysis

4
5
6
 Regression analysis
reportssddsvffffssdn
TABLE OF CONTENTS:
Abstract…………………………………………………………………………
…………………………………3
Introduction……………………………………………………………………
…………………………….….3
Background……………………………………………………………………
………………………………..4
Methodology…………………………………………………………….........
...............................5
Table of rice data of larkana………………………………
…………………………………………..5
Result……………………………………………………………………………
…………………………………6
Discussion………………………………………………………………………
……………………………….7
Graphical
presentation……………………………………………………………………
…….…………8
Conclusion………………………………………………………………………
………………………………8
Recommendation……………………………………………………………
……………………………….9
Reference………………………………………………………………………
……………………..…………9

ABSTRACT:
7
In this report the linear regression analysis technique is used to predict the rice
production. The rice data of District Larkana is collected from Agriculture Statistic
Department, Islamabad with two factors: Area under Cultivation and Production.
The linear regression technique is applied to calculate the relationship between
the Area under Cultivation (Independent) and its effect on Production
(Dependent). The positive, moderate and significant relationship is observed
between the dependent and independent variables. This study can helps to
researchers for knowing the worth of analytics techniques for prediction of
harvest.

INTRODUCTION:
Agriculture is important profession all over the world that serves approximately,
70% humans. Most of the Pakistanis lives in rural areas and are attached with
agribusiness. Agriculture sector is the one of the core part and contributes in
Pakistan economy approximately 23.4 %. It is observed that most of the farmers
uses traditional production method and earn less profit. Pakistan’s economy
mainly relies on four sectors i.e. Food, Fiber crops and Horticulture, Livestock and
Dairy, Fisheries and Forestry. Pakistan produces its most of the economy through
agriculture sector. As per agribusiness indication, the cultivating is engaged in
various process of formation and distribution of agriculture products,
manufacturing of farmstead hardware, equipment and supplies along with their
supervision and management. In order to improve the efficiency and
enhancement of the agribusiness, the research scholars propose modern and
advanced hardware and software technologies. Various statistical techniques
have been used to solve the problems and issues pertaining to agricultural data
analyzing and prediction. These techniques are Sentimental Analysis, Regression
Analysis (RA) Machine Learning (ML), Social network analysis. The Regression
Analysis is used for prediction of rice production of District Larkana. Crop yield
prediction is necessary for rapid decision making. The yield prediction systems are
also available and help to the farmers in reducing the losses. This report
presented the analysis of the rice data of District Larkana using Regression
Analysis technique to predict the total production.

BACKGROUND:

8
  RICE PRODUCTION:
Worldwide, Pakistan has 4th position in the rice export. Approximately, 25% rice
produced in six districts of province Sindh i.e. Thatta, Badin, Dadu, Larkana,
Shikarpur and Jacobabad. Among the all districts, approximately 20% rice
produced in Larkana hence, data of this district is preferred for experiments .

 RICE DATA COLLECTION:

The rice data of District Larkana is mandatory component of this research so that
the statistical data of rice crop is taken from the Agriculture Statistics
Department, Islamabad. It is noted that the data is available at country and
province level. The rice data set of district Larkana is manually prepared from the
year 1981-82 to 2015-16 according to the district-wise percentage Two Talukas of
Larkana i.e. Dokri and Bakrani are mainly considered for the cultivation of the
crop rice. The rice data is collected and prepared with two variables: Area (In
Hectares), and Production (In Million Tons). These variables are used to perform
Regression Analysis on the prepared rice data sets. The collected and prepared
rice data of District Larkana from the year 1981-82 to 2015-16

METHODOLOGY:
Regression is a statistical device specifically designed to compute the average
relationship between the selected dependent and independent variables in terms
of different units of data. The Regression Analysis is a process of fitting any model
or function to given data. The values of dependent variable from the independent
variables can be estimate or predict using this analysis technique. In this study,
the factor Area under Cultivation (AUC) is considered to contribute to rice
production and found the relation and influence of selected factor on the
production of rice. The production is selected as a dependent variable (Y) and
AUC is considered as independent variable (X) to experiment the Regression
Analysis on the rice crop. The relationship between the Area Under Cultivation
and Production is analyzed and calculated using the linear regression model.

DATA ANALYSIS INPUT:

 Rice Data of Larkana from 1981 to 2015:

9
 AREA PRODUCTION
YEAR (In Hectares) (In Million Tons)
(X) (Y)
1981-82 186600 489300
1982-83 177100 447400
1983-84 177300 484100
1984-85 168000 443600
1985-86 168700 414400
1986-87 208200 597400
1987-88 207900 596300
1988-89 208100 597500
1989-90 188800 539200
1990-91 204900 591800
1991-92 205100 585700
1992-93 189300 525000
1993-94 209300 706200
1994-95 191900 583000
1995-96 190400 552900
1996-97 210400 660400
1997-98 211900 630300
1998-99 214400 671500
1999-00 213700 714600
2000-01 162800 566500
2001-02 130800 376900
2002-03 152500 477600
2003-04 170700 515100
2004-05 161800 508800
2005-06 194300 636900
2006-07 81,439 283,351
2007-08 96,200 327,939
2008-09 95,088 370,165
2009-10 97,823 383,600
2011-12 98,935 388,447
2012-13 93,944 362,100
2013-14 98,009 322,418
2014-15 102,134 295,284
2015-16 105,223 409,470

10
 RESULTS:

Regression Statistics
Multiple R 0.91828
R Square 0.843238
Adjusted R
Square 0.838339
Standard Error 48959.2
Observations 34

ANOVA
df SS MS F Significance F
172.131221
Regression 1 4.13E+11 4.13E+11 7 2.01605E-14
Residual 32 7.67E+10 2.4E+09
Total 33 4.89E+11

Coefficient Standard Upper Lower Upper

s Error t Stat P-value Lower 95% 95% 95.0% 95.0%
0.00458499
Intercept 97365.09 31936.14 3.048744 3 32313.31003 162416.9 32313.31 162416.9
AREA 2.466005 0.187959 13.11988 2.01605E-14 2.083144334 2.848866 2.083144 2.848866

DISCUSSION OF RESULTS:
 SIGNIFICANCE OF MODEL:
Anova:
- This Anova table, tests the significance of our regression model.
- The F-statistic of 172.13 indicate that the model as a whole is a good fit for the
data and statistically significant.
- The p-value associated with the F-statistic (2.01605E-14) is very low, providing
strong evidence against the null hypothesis.

11
  COEFFICIENTS:
Intercept (97365.09):
-The intercept represents the value of the dependent variable when all
independent variables are zero. Here, it's approximately 97365.09.

Area (2.466005202):
-It is the coefficient for the independent variable. This coefficient indicates the
change in the dependent variable is approximately 2.47 units for a one unit
increase in the independent variable. It is highly significant with a p-value of
(2.01605E-14).
-We can conclude that the relationship between AREA and the Dependent
variable is statistically significant 97365.09

R Square:
-R square is also known as Coefficient of determination
- R Square indicates the proportion of the variance in the dependent variable that
is predicted by the independent variable(s).
- In this case, approximately 84.32% of the variance in the dependent variable is
predictable by the independent variable (AREA), suggesting that the model
provides a good fit to the data.

GRAPHICAL REPRESENTATION OF RELATIONSHIP OF

VARIABLES

12
 PRODUCTION
800000
Production in Million tons

700000

600000

500000

400000

300000

200000
50000 70000 90000 110000 130000 150000 170000 190000 210000 230000

Area in Hectors

CONCLUSION:
Overall, the regression model suggests that there is a strong positive linear
relationship between the independent variable (AREA) and the dependent
variable. The model explains a significant portion of the variance in the
dependent variable, and the coefficient for AREA is statistically significant.
Therefore, the equation for this regression model would be:

Y = 97365.09 + 2.466 × AREA

This equation quantifies the relationship between the dependent variable and the
independent variable (AREA) based on the coefficients estimated from the
regression analysis.

RECOMMENDATIONS:
1. Use modern technologies:
Adopt advanced hardware and software technologies to increase the efficiency,
growth and enhancement of agribusiness. Encourage continuous research and
development in agribusiness and agriculture to meet daily challenges and avail
opportunities.

2. Farmers' training:

13