FORECASTING APPLE PRODUCTIONS IN INDIA

USING ARIMA MODEL

A Project submitted to
THE MADURA COLLEGE
an autonomous college affiliated to
MADURAI KAMARAJ UNIVERSITY

By
S. SHEHA
(2020PGS015)
In partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
in
STATISTICS

DEPARTMENT OF STATISTICS
THE MADURA COLLEGE
MADURAI- 625011
June 2022

DEPARTMENT OF STATISTICS
 The Madura College (Autonomous)
MADURAI- 625011

Madurai-625011

CERTIFICATE
This is certify that the project work entitled,
“FORECASTING APPLE PRODUCTIONS IN INDIA USING ARIMA
MODEL” done by SNEHA. S, (Reg. No : 2020PGS015) of The Madura
college, Madurai as a partial fulfillment of Post-Graduation in Statistics
during the academic year 2020-2022 is bonafide record of the project work
done under my guidance and supervision.

Signature of the Guide Head of the Department

Viva-voce examination was conducted on--------------- at The Madura College,

Madurai – 625011

Internal Examiner Controller of Examinations External Examiner

 DECLARATION

I hereby declare that the project work entitled, “FORECASTING

APPLE PRODUCTIONS IN INDIA USING ARIMA MODEL” submitted

to The Madura College, Madurai in partial fulfillment of the requirements for

the award of the Degree of MASTER OF SCIENCE IN STATISTICS is a

record of original work done by me under the supervision and guidance of Dr.

P.Vetri Selvi, M.Sc.,M.phil., Ph.D Assistant Professor and Head (i/c),

Department of Statistics, The Madura College, Madurai and that is has not

formed the basis for the award of any Degree / Diploma / Associate Ship /

Fellowship or other similar title to any candidate of any University.

Madurai.
Date:

Signature of the Guide Signature of the Candidate

(P.VETRI SELVI) (SNEHA. S)
.
 ACKNOWLEDEMENT

I thank God almighty for giving me the strength, courage and the
perseverance to pursue the task successfully throughout the preparation of
this project work.

I offer grateful thanks to Dr. J. Suresh, Principal, The Madura College,

Madurai.

I express my heartiest thanks to my guide Dr. P. Vetri Selvi, for his

remarkable guidance during the entire course of this project.

I have greatest pleasure to thank all the respondents for their active
participation. My heartfelt thanks to the staff members of department of
statistics, classmates and well-wisher for their kind support during this project
work

Finally, I owe a lot to my family members and friends for their love and
care during this project work.

SNEHA .S

(2020PGS015)
 CONTENTS

CHAPTER PARTICULARS PAGE No.

No.

1-6
1. INTRODUCTION

2. CLASSIFICATION OF HORTICULTURE 7 - 17

3. METHODOLOGY 18 - 31

4. STATISTICAL ANALYSIS 32 - 41

5. CONCLUSION 42 - 43

6. REFERENCES 44 - 46
 CHAPTER- I

INTRODUCTION

1|P a ge
INTRODUCTION

The horticulture sector encompasses a wide range of crops e.g., fruit crops,

vegetables crops, potato and tuber crops, ornamental crops, medicinal and aromatic

crops, spices and plantation crops. While the first few Five Year Plans assigned priority

to achieving self sufficiency in food grain production, over the years, horticulture has

emerged as an indispensable part of agriculture, offering a wide range of choices to the

farmers for crop diversification. It also provides ample opportunities for sustaining large

number of agro-industries which generate substantial employment opportunities.

With its wide climate and soil variability, India is ideal for growing a wide

variety of horticultural crops such as fruits, vegetables, potatoes, tropical tuber crops,

and mushrooms; ornamental crops; medicinal and aromatic plants, spices, and

plantation crops such as coconut, greennut, cashew, cocoa, tea, coffee, and rubber.

Immediately after gaining independence in 1947, the Government of India placed

a strong focus on establishing self-sufficiency in food production, particularly wheat.

Green Revolution was effectively implemented in the late 1960s and early 1970s. It also

shown that horticulture crops, which are well adapted to the Indian terrain and agro-

climate, might be an excellent alternative for small farmers seeking to achieve

sustainability. However, it was not until the mid-1980s that the Government of India

identified horticulture crops as a means of diversification for making agriculture more

profitable through efficient land use, optimum utilisation of natural resources (soil,

2|P a ge
water, and environment), and creation of skilled employment for rural masses,

particularly women folk.

Previous initiatives have resulted in enhanced output, productivity, and

availability of horticultural crops. As a result, India has become the world's greatest

producer of coconut, arecanut, cashew, ginger, turmeric, black pepper, and tea, as well

as the second largest producer of fruits and vegetables. Among the new crops

successfully introduced for commercial production in the nation are kiwi, olive crops,

and oil palm.

The changing environment stimulates private investment in high-tech

horticulture, including micro-propagation, protected cultivation, drip irrigation,

fertigation, and integrated fertilizer and pest control, as well as the utilization of cutting-

edge post-harvest technologies.

Measures, particularly for perishable items. As a result, horticulture crop

production has shifted from rural to commercial operations, attracting millennials since

it has shown to be intellectually gratifying as well as commercially rewarding.

Horticulture is rapidly being recognized as a sunrise industry, owing to its ability to

increase farm revenue, offer livelihood security, and generate foreign cash through

export. The different agro-climatic conditions, as well as the great diversity of crops and

genetic resources, allow India to produce a wide range of horticulture crops all year.

At the same time of year, India produces a tropical fruit like mango and a

subtropical fruit like apple. Horticulture includes a broad variety of crops such as fruits,

vegetables, flowers, spices, plantation crops such as coconut, drinks such as tea and

3|P a ge
coffee, and certain medicinal and fragrant plants. According to National Horticulture

Development Board figures, the United States accounts for 13% of world fruit output

and 21% of global vegetable production. In both commodity groupings, India is the

second largest producer behind China (Horticultural statistics at a Glance 2017).

1.1 OBJECTIVE OF THE STUDY

The main objective of this study is to develop an ARIMA model for forecasting

different types of Fruit crops productions like Apple, Banana and Guava in the India.

1.2 COLLECTION OF DATA

The National Horticulture Board website was used to collect data for the study.

The data shows the year of production and productivity of India's primary crops.

1.2.1 NATIONAL HORTICULTURE BOARD (NHB)

The National Horticulture Board (NHB) was set up by the Government of India in

1984 as an Autonomous organization under the administrative control of Ministry of

Agriculture and Farmers Welfare and registered as a society under Societies

Registration Act with its headquarters at Gurugram. Presently, NHB has 29 field

offices located all over the country. The broad aims and objectives of the Board are to

develop production clusters/hubs for integrated Hi-tech commercial horticulture,

development of Post-harvest and cold chain infrastructure, ensuring availability of

quality planting material and to promote adoption of new technologies/tools/

techniques for Hi-tech commercial horticulture etc.

The objectives of the Board are as under:-

4|P a ge
1. Development of hi-tech commercial horticulture in identified belts and make such

areas vibrant with horticultural activity, which in turn will act as hubs for

development of horticulture.

2. Development of modern post-harvest management infrastructure as an integral

part of area expansion projects or as common facility for cluster of projects.

3. Development of integrated, energy efficient cold chain infrastructure for fresh

horticulture produce.

4. Popularization of identified new technologies / tools / techniques for

commercialization / adoption, after carrying out technology and need assessment.

5. Assistance in securing availability of quality planting material by promoting

setting up of scion and root stock banks / mother plant nurseries and carrying out

accreditation / rating of horticulture nurseries and need based imports of planting

material.

6. Promotion and market development of fresh horticulture produce.

7. Promotion of field trials of newly developed/imported planting materials and other

farm inputs; production technology; PHM protocols, INM and IPM protocols and

promotion of applied R&D programmes for commercialization of proven

technology.

8. Promotion of Farm Mechanization in Horticulture through demonstration and its

uses at farmers field level to reduce labour cost and increase the productivity of

Horticulture crops.

5|P a ge
9. Promotion of applied R & D for standardizing PHM protocols, prescribing critical

storage conditions for fresh horticulture produce, bench marking of technical

standards for cold chain infrastructure etc.

10. Transfer of technology to producers/farmers and service providers such as

gardeners, nurserymen, farm level skilled workers, operators in cold storages,

work force carrying out post harvest management including processing of fresh

horticulture produce and to the master trainers.

11. Promotion of consumption of horticulture produce and products.

12. Promoting long distance transport solution for bulk movement of horticulture

produce through rail etc.

13. Carrying out studies and surveys to identify constraints and develop short and long

term strategies for systematic development of horticulture and providing technical

services including advisory and consultancy services.

1.3 BRIEF DESCRIPTION OF THE STUDY IN DIFFERENT CHAPTER

Chapter II provides a Classification and Features of Horticulture

Chapter III provides a detailed discussion of the statistical tools and

methodologies used in the project.

Chapter IV covers the statistical analysis findings, interpretation, and

conclusion.

6|P a ge
 CHAPTER- 1I

CLASSIFICATION OF HORTICULTURE

7|P a ge
2.0 INTRODUCTION

Horticulture is a science, as well as, an art of production, utilization and improvement

of horticultural crops, such as fruits and vegetables, spices and condiments,

ornamental, plantation, medicinal and aromatic plants. Horticultural crops require

intense care in planting, carrying out intercultural operations, manipulation of growth,

harvesting, packaging, marketing, storage and processing. India is the second largest

producer of fruits and vegetables in the world after China. In India, about 55–60 per

cent of the total population depends on agriculture and allied activities. Horticultural

crops constitute a significant portion of the total agricultural produce in India. They

cover a wide cultivation area and contribute about 28 per cent of the Gross Domestic

Product (GDP). These crops account for 37 per cent of the total exports of agricultural

commodities from India.

2.1 HORTICULTURE AND ITS IMPORTANCE

The term horticulture is derived from two Latin words hortus, meaning ‘garden’, and

culture meaning ‘cultivation’. It refers to crops cultivated in an enclosure, i.e., garden

cultivation.

2.2 FEATURES AND IMPORTANCE

Horticulture crops perform a vital role in the Indian economy by generating

employment, providing raw material to various food processing industries, and higher

farm profitability due to higher production and export earnings from foreign

exchange.

8|P a ge
 1. Horticulture crops are a source of variability in farm produce and diets.

2. They are a source of nutrients, vitamins, minerals, flavour, aroma, dietary

fibres, etc.

3. They contain health benefiting compounds and medicines.

4. These crops have aesthetic value and protect the environment.

5. The comparative production per unit area of horticultural crops is higher than

field crops, e.g., paddy crop gives a maximum yield of only 30 q/ha, while

banana crop gives 300–450 q/ha and grapes 90–150 q/ha.

6. Fruit and plantation crops can be cultivated in places where the slope of land is

uneven or undulating. Mango and cashew nut are cultivated on a large scale in

hilly and hill back area of the Konkan region.

7. The crops are useful for cultivation in wasteland or poor quality soil.

8. Such crops are of high value, labour intensive and generate employment

throughout the year.

9. Horticultural produce serves as raw material for various industries, such as

processing, pharmaceutical, perfumery and cosmetics, chemical, confectionery,

oils and paints, etc.

10. They have national and international demand and are a good source of foreign

exchange.

2.4 PROSPECTS OF HORTICULTURAL CROPS IN INDIA

Diverse agro-climatic conditions in India ensure the production of all types of fresh

fruits, vegetables and medicinal plants in different parts of the country (Table1.1).

9|P a ge
Health consciousness among people is increasing. Majority of the population in India

is vegetarian. As a result, the demand of fruits and vegetables is also high.

The production of horticultural commodities is far less as compared to the existing

demand in the country. So, there is a vast scope to produce more horticultural crops.

Major areas in the country are suitable only for horticultural crops, like mango, tea,

coconut and arecanut, as they are non-arable, rocky, stony, marshy, undulated and

sloppy. There has been an increase in irrigation facilities but there are crops, which

even with little watering, can survive. One only needs to ensure adequate water

management. Some dry land horticultural crops, like jamun, ber, tamarind, wood

apple, custard apple, ramphal, etc., can be grown on rainfed land also. Compared to

other countries, agricultural labour and other agricultural inputs are far cheaper and

easily

This leads to the use of intensive methods and improved technology in the production

of horticultural crops. Awareness of storage and processing methods also increase the

availability of the produce, job opportunity and income generation.

2.5 BRANCHES OF HORTICULTURE AND SPECIAL HORTICULTURAL

OPERATIONS

Horticulture is perhaps the most important branch of agriculture. It is further divided

into four different branches.

2.5.1 POMOLOGY

10 | P a g e
 The term is derived from Latin words poma and logus. Poma means ‘fruit’ and

logus means ‘study, knowledge or discourse’. It can be defined as a branch of

horticulture, which deals with the scientific study of fruit crops.

2.5.2 OLERICULTURE

The term is derived from Latin words olerus meaning ‘vegetables’ and cultura

meaning ‘cultivation’. It can be defined as a branch of horticulture, which deals with

the scientific study of vegetable crops.

2.5.3 FLORICULTURE

The term floriculture is derived from Latin words florus and cultura. Florus

means ‘flower’ and cultura means ‘cultivation’. It can be defined as a branch of

horticulture, which deals with the scientific study of flowering and ornamental crops

Landscaping is the art of beautifying a piece of land using garden designs, methods

and plant material. Professionals who do landscaping are called ‘landscape architects’.

2.5.4 POST-HARVEST TECHNOLOGY

It is a branch of horticulture, which deals with the principles and practices of

handling, packaging and processing of harvested crops to increase their storage life

and availability.

2.6 CLASSIFICATION OF HORTICULTURAL CROPS

Many varieties of horticultural crops are produced across India. India is gifted with a

wealth of plants. Comprised of 356 domesticated species of economic relevance and

326 wild forms or relatives. Aside from that, the Indian subcontinent has a wide

11 | P a g e
diversity of plant richness; around 9,500 additional species of ethanobotanical

importance have been reported.

Out of these more than 50 types of individuals types of species, plantation crops, etc.

are under commercial cultivation in different part of the country under different sets

of growing conditions. An attempt to deal with all these plants separately becomes

tedious, cumbersome and infeasible and more so repetitive. To avoid these difficulties

it is better to classify the plants in groups based on similarity in either of the traits are

placed under one group. Such type of grouping of plants in different categories is

referred to as classification. To overall objectives of the classification is to systemize

the presentation and make the remembrance of the plants easy and convenient.

2.7 IMPORTANCE OF HORTICULTURE CROPS

Horticulture enriches diets:horticuiture –specifically,growing fruits and vegetables-

provides critical nutrients for a balanced diet.Diet low in fruits and vegetables

contribute significantly to some of the worlds most widespread and debilitating

nutrient-related disorder.

Color,texture,size,and flavor continue to be the predominant quality characteristics

important for successful international marketing of horticultural crops.

Product nutraceutical value is also a quality characteristic playing a significantly more

important role among consumers in making purchasing decisions.

12 | P a g e
FRUITS

A large variety of fruits are grown in India. Of these mango, banana, citrus, pineapple,

papaya, guava, sapota, jackfruit, litchi and grape, among the tropical and sub-tropical

fruits; apple, pear, peach, plum, apricot, almond and walnut among the temperate

fruits and aonla, ber, pomegranate, annona, fig, phalsa among the arid zone fruits are

important.

MANGO

Mango is the most important fruit covering 37.60 per cent of area and accounting

for22.21 per cent of total fruit production in the country. India’s share in the world

production of mango is 54.2 per cent. The area and production of mango (1998-99)

was 1.401 million ha with a production of 9.782 million tonnes. Major mango

producing states are Andhra Pradesh, Bihar, Karnataka, Maharashtra and U.P. Andhra

Pradesh ranks first in mango production with a share of 20 percent and highest in

productivity. There has been an increase of 29.62 and 11.77 per cent in area and

production respectively in mango between 1991-92 and 1998-99. Comparatively

lesser increase in production seems to be due to larger area under new plantations

which are yet to come to full bearing stage.

CITRUS

Citrus ranks second in total area with 0.488 million ha and 13.09 percent area under

fruits with production of 4.575 million tonnes (10.39 per cent of total production

under fruits). Limes, lemons, sweet orange and mandarin cover bulk of the area under

this group of fruits. Cultivation of grapefruit and pummelo introduced decades back
13 | P a g e
did not catch up commercially. Citrus fruits are grown mainly in the states of

Maharashtra, Andhra Pradesh, Punjab, Karnataka and N.E. region. The increase in

area and production of citrus in India (1991-92 to 1998-99) has been of the order of

28.20 and63.12 per cent respectively. Productivity has increased from 7.21 to 9.40

t/ha (1991-92 to 1998-99). Area under Kogazi-lime has increased significantly all

over the country

BANANA

Banana ranks third in area with 0.464 million ha covering 12.46 per cent of the total

area. However, it is first in total production (15.07 million tonnes), being nearly one-

third (34.22 per cent) of total fruit production. India occupies first position in banana

production globally. Among the states, Tamil Nadu ranks first in area and production

while productivity is highest in Maharashtra. Most of the banana is produced on a

small scale basis in different production systems. Total increase in area and

production of banana has been 21.05 per cent and 93.83 per cent respectively

GUAVA

Guava is the fifth important fruit covering an area of 0.151 million ha with a total

production of 1.8 million tonnes. This fruit accounts for 4.06 and 4.09 per cent of the

total area and production, respectively. The increase in area and production in guava

between 1991-92 and 1998-99 has been 66.66 and 63.63 per cent respectively.

14 | P a g e
PAPAYA

Papaya ranks sixth in area and production with 0.068 million ha (1.82 per cent) of the

total area under fruits and 1.582 million tonnes (3.59 per cent) of the and 1998-99 has

been 40.00 and 97.00 per cent respectively. The phenomenal increase both in area and

production has been due to the development of several gynodioecious and high quality

varieties in different states of the country to adoption of high density planting, use of

tissue-cultured seedlings and drip irrigation, which significantly improved

productivity.

APPLE

Apple Is the fourth major fruit crop of the country and occupies a total area of

0.231 million ha with production of 1.38 million tonnes. It is grown mainly in the

states of Jammu and Kashmir, Himachal Pradesh, Uttaranchal and Arunachal Pradesh.

Share of total area and production of apple apple in the total fruit production during

1998-99 was 6.2 and 3.13 per cent, respectively.

GRAPE

Another fruit in which significant increase in area and production has taken place is

grapes. Grape occupies 1.14 per cent of the total area with 2.56 per cent of the total

production of fruits. The total area and production during 1998-99 were 0.043 million

ha and 1.083 million tonnes respectively. Though primarily grown in Maharashtra,

Andhra Pradesh, Karnataka and Tamil Nadu, it is also grown on a limited area in the

15 | P a g e
plains of northern states. The area and production increases (1991-92 to 1998-99)

have been 33.33 and 61.19 per cent respectively.

LITCHI

Litchi is assuming importance owing to its high economic returns and export

potential. It occupies 1.5 per cent of the total area and 0.97 per cent of total production

of fruits (1998-99). It is largely grown in Bihar and has potential for its expansion in

sub- tropical regions free from hot winds during April-May. It is grown in small

pockets in Punjab, Haryana, U.P., West Bengal and N.E region. The increase in area

and production of litchi (1991-92 to 1998-99) has been 20.0 and 79.16 per cent

respectively.

2.7 HORTICULTURE SECTOR

The need for diversification to horticulture sector was acknowledged by the

Government of India in mid-eighties by focusing its attention on investment in this

sector. Presently horticulture has established its credibility in improving income

through increased productivity, generating employment and in enhancing exports.

Resultantly, horticulture has moved from rural confines to commercial venture.

The Department of Agriculture, Co-operation & Farmers Welfare (DAC&FW) of the

Ministry of Agriculture & Farmers Welfare is the nodal department for over viewing

horticulture development in the country. It implements different programmes through

Departments of Horticulture/Agriculture in all the States and provides the leadership

to coordinate activities for the promotion of horticulture. However, efficient

implementation of programme and policies call for robust information in time. In this

16 | P a g e
backdrop, this book presents the overall information on horticulture focusing area,

production and yield estimates apart from the information about crops price, market

arrivals and export.

The scenario of horticulture crops in India has become very encouraging. The

percentage share of horticulture output in Agriculture has become 33%. Under the

purview of Agriculture & allied activities, the share of plan outlay for Horticulture

which was 3.9% during IX Plan, has increased to 4.6% during the XII Plan.

17 | P a g e
 CHAPTER- III

METHODOLOGY

18 | P a g e
 METHODOLOGY

3.1 TYPES OF DATA

A variate, often known as a random variable, is a quantity or trait whose value varies

from one unit of research to the next. For example, the units may be headache

sufferers, and the variate could be the time between taking an aspirin and the headache

going away. A variate's observation or reaction is the value it takes for some given

unit. There are various types of variate.

 Nominal or qualitative; described by a word or phrase (e.g. blood group, colour).

 Quantitative; described by a number (e.g. time till cure, number of calls arriving at a

telephone exchange in 5 seconds).

 Ordinal; this is an "in-between" case. although observations are not numbers, they

may be arranged (e.g. much improved, improved, same, worse, much worse).

Averages and the like can be evaluated rationally for quantitative data, but not for the

other two. The frequencies of distinct categories can be used to analyse qualitative

data. Ordinal data may be analysed similarly to qualitative data, but it requires

specialised approaches known as nonparametric methods.

Quantitative data can be:

 Discrete: the variate can only take one of a finite or countable number of values (e.g.

a count)

19 | P a g e
 Continuous: the variate is a measurement which can take any value in an interval of

the real line.

3.2 DESCRIPTIVE STATISTICS

3.2.1 Central tendency

Measures of central tendency are the center values of a data set.

 Mean is the average of all the data. Its symbol is 𝑥̅ .

 Mode is the data value appearing most often in the data set.

 Median is the middle value of the data set, arranged in ascending order.

3.2.2 Measures of Dispersion

In statistics, measures of dispersion describe how spread apart the data is from the

measure of center. There are three main types of dispersion:

 Variance - the mean of the squares of the distance each data item (𝑥𝑖 ) is from

the mean.

𝟐
(𝑥𝟏 − 𝑥̅ )𝟐 + (𝑥𝟐 − 𝒙
̅)𝟐 + ⋯ + (𝑥𝑛 − 𝑥̅ )𝟐𝑛
𝜎 =
𝑛

The symbol for variance is 𝜎 𝟐

 Standard deviation - the square root of the variance.

Standard deviation measures how closely the data clusters around the mean. It is the

square root of the variance. Its symbol is σ.

20 | P a g e
 (𝑥𝟏 − 𝑥̅ )𝟐 + (𝑥𝟐 − 𝒙
̅)𝟐 + ⋯ + (𝑥𝑛 − 𝑥̅ )𝟐𝑛
𝜎=√
𝑛

 Range - the difference between the highest and lowest values in the data.

3.3 LINE CHART

A line chart or line plot or line graph or curve chart is a type of chart which displays

information as a series of data points called 'markers' connected by

straight line segments. It is a basic type of chart common in many fields. It is similar

to a scatter plot except that the measurement points are ordered (typically by their x-

axis value) and joined with straight line segments. A line chart is often used to

visualize a trend in data over intervals of time – a time series – thus the line is often

drawn chronologically. In these cases they are known as run charts.

3.4 AUGMENTED DICKEY–FULLER TEST

In statistics and econometrics, an augmented Dickey–Fuller test (ADF) tests the null

hypothesis that a unit root is present in a time series sample. The alternative

hypothesis is different depending on which version of the test is used, but is

usually stationarity or trend-stationarity. It is an augmented version of the Dickey–

Fuller test for a larger and more complicated set of time series models.

21 | P a g e
The augmented Dickey–Fuller (ADF) statistic, used in the test, is a negative number.

The more negative it is, the stronger the rejection of the hypothesis that there is a unit

root at some level of confidence.

The testing procedure for the ADF test is the same as for the Dickey–Fuller test but it

is applied to the model

Δ𝑦𝑡 = 𝛼 + 𝛽𝑡 + 𝛾𝑦𝑡−1 + δ1 Δ𝑦(𝑡−1) + ⋯ … . . δp Δ𝑦(𝑝−1 ) +∈𝑡

where is 𝛼 constant, 𝛽 the coefficient on a time trend and the lag order of the

autoregressive process. Imposing the constraints 𝛼 = 0 and , 𝛽 = 0 corresponds to

modelling a random walk and using the constraint , 𝛽 = 0 corresponds to modeling a

random walk with a drift. Consequently, there are three main versions of the test,

analogous to the ones discussed on Dickey–Fuller test (see that page for a discussion

on dealing with uncertainty about including the intercept and deterministic time trend

terms in the test equation.)

By including lags of the order p the ADF formulation allows for higher-order

autoregressive processes. This means that the lag length p has to be determined

when applying the test. One possible approach is to test down from high orders

and examine the t-values on coefficients. An alternative approach is to examine

information criteria such as the Akaike information criterion, Bayesian

information criterion or the Hannan–Quinn information criterion.

The unit root test is then carried out under the null hypothesis 𝛾 = 0against the

alternative hypothesis of 𝛾 < 0. Once a value for the test statistic

22 | P a g e
 𝛾̂
𝐷𝐹𝛾 =
𝑆𝐸 (𝛾̂ )

is computed it can be compared to the relevant critical value for the Dickey–

Fuller test. As this test is asymmetrical, we are only concerned with negative

values of our test statistic 𝐷𝐹𝑟 . If the calculated test statistic is less (more

negative) than the critical value, then the null hypothesis of 𝛾 = 0 is rejected and

no unit root is present.

3.5 AUTOREGRESSIVE INTEGRATED MOVING AVERAGE

(ARIMA)

In the extension of ARMA model to deal with homogenous non-stationary time series

in which 𝑋𝑡 , itself is non-stationary but its 𝑑 𝑡ℎ difference is a stationary ARMA

model. Denoting the 𝑑 𝑡ℎ difference of 𝑋𝑡 , by

𝜑(𝐵) = 𝜙 (𝐵)𝛻 𝑑 𝑋𝑡 = 𝜃 (𝐵)𝜀𝑡

where 𝝋(𝑩) is the nonstationary autoregressive operator such that d of the roots

of 𝝋(𝑩) = 𝟎 are unity and the remainder lie outside the unit circle. 𝝓(𝑩) is a

stationary autoregressive operator

23 | P a g e
3.5.1 Stationarity
The foundation of time series analysis is stationarity. Consider a finite set of return

variables {𝑅𝑡1 , 𝑅𝑡2 , ⋯ , 𝑅𝑡𝑛 } from a time series process, {𝑅(𝑡): 𝑡 = 0, +1, +2, ⋯ }. The

k-dimensional distribution function is defined as

𝐹𝑅 𝑡1 (𝑟1 , 𝑟2 , ⋯ , 𝑟𝑘 ) = 𝑃{𝑅𝑡1 ≤ 𝑟1 , 𝑅𝑡2 ≤ 𝑟2 , ⋯ , 𝑅𝑡𝑘 ≤ 𝑟𝑘 },

⋯𝑅𝑡𝑘

where 𝑟𝑗 , 𝑗 = 1,2, ⋯ , 𝑘 are any real numbers.

A process is said to be:

1) first-order stationary in distribution if its one-dimensional distribution is time

invariant. That is, if

𝐹𝑅 𝑡1 (𝑟1) = 𝐹𝑅 𝑡(1 +𝑘) (𝑟1)

for any integers 𝑡1 , 𝑘 and 𝑡1 + 𝑘

2) second-order stationary in distribution if

𝑭𝑹 𝒕𝟏…..𝑹 (𝒓𝟏 , 𝒓𝟐 ) = 𝑭𝑹 𝒕𝟏+𝒌…..𝑹 +𝑲 (𝒓𝟏 , 𝒓𝟐 )

𝒕. 𝒕 .

for any integers 𝑡1 , 𝑡2 , 𝑘, 𝑡1 + 𝑘 and 𝑡2 + 𝑘.

3) the 𝑛𝑡ℎ -order stationary in distribution if

𝑭𝑹 𝒕 (𝒓𝟏 … … , 𝒓𝒏 ) = 𝑭𝑹 𝒕𝟏+𝒌…..𝑹 +𝑲,𝑹 (𝒓𝟏 , … … 𝒓𝒏 )

𝟏 …..𝑹𝒕𝒏 _ . 𝒕 𝒕 𝒏 +𝒌 .

for any (𝑡1 , ⋯ , 𝑡𝑛 ) and k integers.

A process {𝑅𝑡 } is weakly stationary if the mean 𝐸(𝑅𝑡 ) = 𝜇 is a fixed constant

for all t and the auto covariance 𝐶𝑜𝑣(𝑅𝑡 , 𝑅𝑡+𝑘 ) = 𝛾𝑘 depends only on the time

24 | P a g e
difference or time lag k for all t.Stationary in the wide sense or covariance stationary

is also referred to as second-order stationary process.

3.5.2 White Noise Process

A process {𝑎𝑡 } is called a white noise process if it is a sequence of uncorrelated

random variables from a fixed distribution with constant mean, 𝑬(𝒂𝒕 ) = 𝝁𝒂 , usually

assumed to be zero, constant variance, 𝑉𝑎𝑟(𝑎𝑡 ) = 𝜎𝑎2 and 𝛾𝑘 = 𝐶𝑜𝑣(𝑎𝑡 , 𝑎𝑡 +𝑘 = 0,

for all k≠0. It is denoted by 𝑎𝑡 ~𝑊𝑁(0, 𝜎𝑎2 ), where WN stands for white noise . By

definition, a white noise process {𝑎𝑡 } is stationary with auto covariance function,

𝜎𝑎2 , 𝑘=0
𝛾𝑘 = {
0, 𝑘≠0

The autocorrelation function is given as:

1, 𝑘 − 0
𝜌𝑘 = {
0, 𝑘 ≠ 0

while the partial autocorrelation function is

1, 𝑘 − 0
𝜑𝑘𝑘 = {
0, 𝑘 ≠ 0

Thus, the implication of a white noise specification is that the ACF and PACF are

identically equal to zero.

3.5.4 Auto covariance and Autocorrelation Functions

Covariance between 𝑅𝑡 and 𝑅𝑡+𝑘 denoted by 𝐶𝑜𝑣 (𝑅𝑡 , 𝑅𝑡+𝑘 ), which is a function of the

time difference, k, is called the autocovariance function {𝛾𝑘 } of the stochastic process.

As a function of k, 𝛾𝑘 is called the autocovariance function in the time series analysis

25 | P a g e
since it represents the covariance between 𝑅𝑡 and 𝑅𝑡+𝑘 from the same process. It is

defined as

𝛾𝑘 = 𝐶𝑜𝑣(𝑅𝑡 , 𝑅𝑡+𝑘 = 𝐸(𝑅𝑡 − 𝜇)(𝑅𝑡+𝑘 − 𝜇)

The sample estimate of 𝛾𝑘 is 𝐶𝑘 given by

𝑛−𝑘
1
𝐶𝑘 = ∑(𝑅𝑡 − 𝑅̅) (𝑅𝑡+𝑘 − 𝑅̅)
𝑛
𝑡=1

Similarly, the correlation between 𝑅𝑡 and 𝑅𝑡+𝑘 denoted by 𝐶𝑜𝑟𝑟(𝑅𝑡 , 𝑅𝑡+𝑘 ,) which is a

function of the time difference, k, is called the autocorrelation function {𝜌𝑘 } of the

stochastic process. As function of k, 𝜌𝑘 is called the autocorrelation function in time

series analysis since it represents the correlation between 𝑅𝑡 and 𝑅𝑡+𝑘 from the same

process. It is defined as

𝐶𝑜𝑣 (𝑅𝑡 , 𝑅𝑡+𝑘 ) 𝛾𝑘

𝜌𝑘 = =
√𝑉𝑎𝑟(𝑅𝑡 )√𝑉𝑎𝑟(𝑅𝑡 +𝑘 ) 𝛾0

The corresponding sample estimate is given by

𝐶𝑘
𝜌̂𝑘 = , 𝑘 = 0,1,2, ⋯
𝐶0

3.5.4. Partial Autocorrelation Function (PACF)

The conditional correlation between 𝑅𝑡 and 𝑅𝑡+𝑘 after their mutual linear dependency

on the intervening variables (𝑅𝑡+1 , 𝑅𝑡+2 , ⋯ , 𝑅𝑡+𝑘−1 has been removed, given

by 𝐶𝑜𝑟𝑟(𝑅𝑡 , 𝑅𝑡+𝑘 /𝑅𝑡+1 , 𝑅𝑡+2 , ⋯ , 𝑅𝑡+𝑘−1 , is usually referred to as the partial

autocorrelation in time series analysis

26 | P a g e
 Partial autocorrelation can be derived from the regression model, where the

dependent variable, 𝑅𝑡+𝑘 , from a zero-mean stationary process is regressed on k-

lagged variables 𝑅𝑡+𝑘−1 , 𝑅𝑡+𝑘−2 , ⋯ and 𝑅𝑡 , that is

𝑅𝑡+𝑘 = 𝜑𝑘 1 𝑅𝑡+𝑘−1 + 𝜑𝑘 2 𝑅𝑡+𝑘−2 + ⋯ + 𝜑𝑘𝑘 𝑅𝑡+𝛼𝑡+𝑘 ,

where 𝜑𝑘𝑖 denotes the 𝑖 𝑡ℎ regression parameter and 𝑎𝑡+𝑘 is an error term with mean

zero and uncorrelated with 𝑅𝑡+𝑘−𝑗 , for j=1,2,⋯,k. Multiplying 𝑅𝑡+𝑘−𝑗 on both sides

of the above regression equation and taking the expectation, we get

𝛾𝑗 = 𝜑𝑘1 𝛾𝑗−1 + 𝜑𝑘2 𝛾𝑗−2 + ⋯ + 𝜑𝑘𝑘 𝛾𝑗−𝑘

Hence,

𝜌𝑗 = 𝜑𝑘1 𝜌𝑗−1 + 𝜑𝑘2 𝜌𝑗−2 + ⋯ + 𝜑𝑘𝑘 𝜌𝑗−𝑘

AR(2) Process

The AR(2) process

𝑋𝑡 = 𝜑1𝑋𝑡 − 1 + 𝜑2𝑋𝑡 − 2 + 𝜀𝑡

may be written in terms of the lag

operator as

(1 — 𝜑1 𝐿 — 𝜑2 𝐿2 )𝑋𝑡 =

𝜀𝑡
We may write the process as

𝑋𝑡 = 𝜓(𝐿)𝜀𝑡

= (1 + 𝜓1 𝐿1 + 𝜓2 𝐿2 + . . . )𝜀𝑡

where

(1 — 𝜑1 𝐿 — 𝜑2 𝐿2 ) − 1 ≡ (1 + 𝜓1 𝐿1 + 𝜓2 𝐿2 + . . . )

or equivalently

27 | P a g e
 (1 — 𝜑1 𝐿1 — 𝜑2 𝐿2 )(1 + 𝜓 𝐿
1 1
+ 𝜓2 𝐿2 + . . . ) ≡ 1

Equating coefficients we get:

𝐿1 : − 𝜑1 + 𝜓1 = 0 𝜓1 = 𝜑1
𝐿2 : − 𝜑2 + φ1 ψ1 + ψ2 = 0 ψ2 = φ1 + φ2
𝐿3 : −𝜑2 + 𝜑1 𝜓1 + 𝜓2 = 0 𝜓3 = 𝜑3 + 2𝜑1 𝜑2
………
𝐿𝑗 : 𝜓𝑗 = 𝜑1 𝜓𝑗−1 + 𝜓2 𝜓𝑗−2
and all weights can be determined recursively.

The AR(1) process was stationary if |α| < 1. What conditions should
we impose on the AR(2) process

(1 — 𝜑1 𝐿1 — 𝜑2 𝐿2 )𝑋𝑡 = 𝜀𝑡

in order that it be stationary? Consider the reciprocals (say g1 and

g2) of the roots of

(1 — 𝜑1 𝐿1 — 𝜑2 𝐿2 ) = 0

Then the equation may be written (1 — 𝑔1 𝐿)(1 — 𝑔2 𝐿) = 0

The process is stationary if |g1| < 1 and |g2| < 1. These roots may be real

or

complex. (It is usually said that |g1|−1 and |g2|−1 lie outside the

unit circle). These restrictions impose the following conditions on

φ1 and φ2.

φ1 + φ2 < 1
—φ1 + φ2 < 1
—1 < φ2 < 1

The ACF (autocorrelation function) of a stationary AR(2) process

may be derived as follows: Multiply the basic equation

28 | P a g e
 𝑋𝑡 − 𝜑1 𝑋𝑡−1 − 𝜑2 𝑋𝑡−2 = 𝜖𝑡

by Xt−k and take expectations

𝐸[𝑋𝑡 𝑋𝑡−𝑘 ] — 𝜑1 𝐸[𝑋𝑡−1 𝑋𝑡−𝑘 ] — 𝜑2 𝐸[𝜀𝑡 𝑋𝑡−𝑘 ] = 𝐸[𝑋𝑡 − 𝜀𝑡 −𝑘 ]

or in terms of autocorrelations,
𝜌𝑘 — 𝜑1 𝜌𝑘−1 — 𝜑2 𝜌𝑘−2 = 0 k = 1, 2 . . .

The observant reader will notice that the autocorrelations obey the
same difference equation as the time series apart from the missing
random term (the corresponding homogeneous difference
equation) and the initial conditions (ρ0 = 1, ρ−1 = ρ1) We can solve
this problem by direct substitution.

For k = 1
ρ1 — φ1ρ0 — φ2ρ−1 = 0 , ρ0 = 1 , ρ1 = ρ−1, 𝜌1 = 𝜑1 ÷ 1 − 𝜑2

For k = 2
𝜑2
ρ2 = 𝜑1 𝜌1 + 𝜑2 𝜌0 = + 𝜑2
1−𝜑2
and all other values may be derived from the recursion and may be
seen to be time independent. We now work out the variance of an
AR(2) system. Put k = 0 in recursion for covariance:
γ0 — φ1γ−1 — φ2γ−2 = σ
ε
2

γ0(1 — φ1ρ1 — φ2ρ2) = σ2

which is independent of t. The conditions on g1 and g2, given

earlier, ensure that 0 < γ0 < ∞.

Thus an AR (2) process is stationary.

The properties of the Autocorrelation function may be derived

from the general solution of the difference equation

𝜌𝐾 — 𝜑1 𝜌𝐾−1 — 𝜑2 𝜌𝐾−2 = 0

29 | P a g e
 which is of the form

𝜌𝑘 = 𝐴𝑔𝑘 + 𝐵𝑔𝑘

where A and B are constants determined by initial conditions ρ0 =

1 and ρ−1 = ρ+1 . If g1 and g2 are real the autocorrelogram is a
mixture of two damped exponentials (i.e. both die out
exponentially). This is similar to a weighted sum of two AR (1)
processes.

If g1 and g2 are complex the ACF is a damped sine wave.

If g1 = g2 the general solution is given by

𝜌𝑘 = (𝐴1 + 𝐴2𝑘 ) 𝑔𝑘

MA Process

An MA(1) process is defined by:

Xt = εt + θεt−1

where εt is white noise

E[Xt] = 0

var[Xt] = E[(εt + θεt−1)2]

= (1 + θ2)σ2

λ1 = E[xtxt−1]

= E[(εt + θεt−1)(εt−1 + θεt−2)]

= θσε2
𝜃
Therefore 𝜌1 =
1−𝜃2

λ2 = E[(εt + θεt−1)(εt−2 + θεt−3)]

= 0
Clearly λj = 0 for j ≥ 2. Thus an MA(1) process is stationary

30 | P a g e
(regardless of the value of θ). An MA(q) process is defined as
follows. εt is as usual a Gaussian White noise.
Xt = εt + θ1εt−1 + · · · + θqεt−q
E[Xt] = 0

var[Xt] = (1 + θ12 + · · · +q θ2)σ

ε
2

λk = Cov[XtXt−k]

= E[(εt + θ1εt−1 + .

. . θkεt−k

θk+1εt−k−1 + ·

· · + θqεt−q )

= (θk + θk+1 θ1 + · · · + θqθq−k)σ2

ε

and 𝜌𝑘 = 𝜆𝑘 \𝑣𝑎𝑟 [𝑋𝑡 ]

31 | P a g e
 CHAPTER- IV

STATISTICAL ANALYSIS

32 | P a g e
 STATISTICAL TECHNIQUES FOR ANALYSIS

In this chapter, we provide a detail discussion of statistical techniques that are used in

the following analysis.

Descriptive Statistics

Minimum Maximum Mean Std. Deviation

Fruits Production in India 0

4332 1163.25 1433.018
(2015-2016)

Fruits Production in India 0

4673 1217.13 1537.891
(2016-2017)

Fruits Production in India 0

5003 1230.25 1585.675
(2017-2018)

Table: 2 Descriptive Statistics of Fruits Production in India

The Table:2 shows that in 2015-2016, fruit production in India varied between 0 and

4332, with a mean standard deviation of 1163.25, 1433.018. In 2016-2017, India's

fruit production varied between 0 and 4673, with a mean and standard deviation of

1217.13, 1537.891, while in 2017-2018, it ranged between 0 and 5003, with a mean

and standard deviation of 1230.25, 1585.675.

33 | P a g e
 Minimum Maximum Mean Std. Deviation

APPLE 1158 2891 1983.71 447.768

GRAPES 881 2922 1945.81 667.920

GUAVA 1683 4054 2657.17 929.413

SAPOTA 594 1744 1228.27 260.785

MANGO 8716 19506 1.46E4 3258.188

BANANA 13304 30808 2.40E4 6355.066

Table: 1 Descriptive Statistics of Major Fruits

The Table: 1 shows that from 2001 to 2018, Apple's output in India fluctuated

between 1158 and 2891, with a mean standard deviation of 1983.71, 447.768. From

2001 to 2018, grape output in India fluctuated between 881 and 2922, with a mean

standard deviation of 1945.81, 667.920. From 2001 to 2018, guava output in India

fluctuated between 1683 and 4054, with a mean standard deviation of 2657.17,

929.413. From 2001 to 2018, sapota output in India fluctuated between 594 and 1744,

with a mean standard deviation of 1228.27, 260.785. From 2001 to 2018, mango

output in India fluctuated between 8716 and 19506, with a mean standard deviation of

1.46E4, 3258.188. From 2001 to 2018, banana output in India fluctuated between

13304 and 30808, with a mean of 2.40E8, 6355.066.

34 | P a g e
 4500
4000
3500
3000
Production

2500
Apple
2000
Grapes
1500
Guava
1000
Sapota
500
0

Year

Figure: 1

Figure 1 shows the production of Apple, Grapes, guava ad Sapota in India

during 2001-2018. Based on the graph above, we can deduce that in 2009-2010, the

production of grapes was decreased and gradually increased after 2010 and also we

noted that the production of guava was gradually increased from the year 2007. We

can deduce that the production of Apple is increase in the year 2010-2011 and after

that it was fluctuated. There is gradually increase in production of Sapota from the

year 2001

35 | P a g e
 6000

5000

4000
Production

3000

2000

1000 2015-2016

0 2016-2017
HIMACHAL PRADESH
ASSAM

JHARKHAND

MAHARASHTRA

PUNJAB
RAJASTHAN
MADHYA PRADESH

WEST BENGAL
MIZORAM
NAGALAND
ODISHA

UTTAR PRADESH
BIHAR

TAMIL NADU
GUJARAT
ARUNACHAL PRADESH

KERALA

MEGHALAYA
ANDHRA PRADESH

CHHATISGARH

TRIPURA
MANIPUR

SIKKIM
KARNATAKA

TELANGANA
2017-2018

State/Uts

Figure 2

Figure 2 shows the production of fruits in India during 2015-2018. Based on the graph

above, we can deduce that in 2015-2016, TamilNadu has the highest production of

fruits and Rajasthan has the fewest. Andhra Pradesh will have the highest production

in 2016-2017, while Himachal Pradesh and Rajasthan will have the fewest. In 2017-

2018, Andhra Pradesh will have the highest production, while Himachal Pradesh will

have the fewest.

36 | P a g e
 FORECASTING OF PRODUCTIVITY OF APPLE BY
FITTING APPROPRIATE ARIMA MODEL

Figure: 3 (a) Plot of original values of Apple Production

By Augmented Dickey -Feller test, ADF Statistics =-2.6264 and p-value=0.0877, this

means the time series is non-stationary. In other words, it has some time-dependent

structure and does not have constant variance over time. The original data plot of

productivity of Apple as shown in Figure3 (a), data is non-stationary.

Figure: 3 (b) Plot of first Differenced of Apple Production

Thus, the first difference of the data on area is plotted and shown in Figure3 (b) this
plot shows that the first differences data is found to be stationary.

37 | P a g e
 Figure:4 (a) ACF Plot of 1st Differenced Apple

Figure:4 (b) PACF Plot of 1st Differenced Apple

The ACF and PACF plot of first difference values of Apple productivity is shown in

Fig. 4 (a)&(b), The correlogram of autocorrelation function (ACF) of differenced

series (Fig. 4) indicates that the auto correlation function falls immediately after one

lag, hence the value of ‘q’ was decided to be ‘0’. Further, the correlogram of partial
38 | P a g e
autocorrelation function (PACF) of differenced series (Fig. 4) indicates that the auto

correlation function falls quickly after the lag 2, hence the value of ‘p’ was decided to

be ‘2’. Thus, the ARIMA (2,1,0) model was selected for parameter estimation, model

validation and forecasting of Apple production in India.

Parameter Estimates for Apple production in India

After identifying the suitable ARIMA (2, 0,1) structure, the parameters of identified

model were assessed and are presented in Table 1(significant at 5% level.)

Model
Estimate SE t Sig.
Parameter

33.526 2.160 .050

Intercept 72.420

AR(1) -.852 .225 -3.792 .002

AR(2) -.560 .228 -2.451 .029

\
Table 1 Estimates of ARIMA (2,1,0) model for Apple production in India

39 | P a g e
Forecast of Apple production in India

After the model identification, estimation of its parameters, the forecasting of

was made to know the future values of Apple production in India by using the selected

and fitted ARIMA model (2,1,0). The forecasted values of Apple production in India

for the year 2019 to 2025 are presented in Table 2.

Year Apple Forecast(Million 95% confidence limits

production Tones)
forecast
Upper value Lower value

2019 2506 3175 1838

2020 2606 3306 1906

2021 2744 3565 1922

2022 2745 3586 1905

2023 2841 3724 1959

2024 2933 3876 1990

2025 2976 3946 2006

Table 2 Forecasting of Apple production with confidence intervals

40 | P a g e
 The graphical representation of actual and fitted values of Apple production by

ARIMA (2,1,0) model along with the forecasted values with 95 percent confidence

interval is also presented in Fig. 5.

Figure 5
Finally, considering all of the Graphical and Formal test, it is obvious that our fitted

model ARIMA (2, 1,0) is the best fitted model for forecasting Apple production in

India.

41 | P a g e
 CHAPTER- V

CONCLUSION

42 | P a g e
CONCLUSION

 The production of Apple, Grapes, guava ad Sapota in India during 2001-2018.

Based on the graph above, we can deduce that in 2009-2010, the production of

grapes was decreased and gradually increased after 2010 and also we noted that

the production of guava was gradually increased from the year 2007. We can

deduce that the production of Apple is increase in the year 2010-2011 and after

that it was fluctuated. There is gradually increase in production of Sapota from the

year 2001.

 We can conclude that in 2015-2016, TamilNadu has the highest production of

fruits and Rajasthan has the fewest. Andhra Pradesh will have the highest

production in 2016-2017, while Himachal Pradesh and Rajasthan will have the

fewest. In 2017-2018, Andhra Pradesh will have the highest production, while

Himachal Pradesh will have the fewest.

 Apple manufacturing in India is expected to be 2506 million tonnes in 2019, 2606

million tonnes in 2020, 2744 million tonnes in 2021and 2745 million tonnes in

2022, 2841 million tonnes in 2023, 2933 million tonnes in 2024 and 2976 million

tonnes in 2025.

43 | P a g e
REFERRENCES

44 | P a g e
References

1. Mohammed Amir Hamjah (2014), Forecasting Major Fruit Crops Productions

in Bangladesh using Box-Jenkins ARIMA Model Journal of Economics and

Sustainable Development, ISSN 2222-1700 (Paper) ISSN 2222-2855 , Vol.5,

No.7.

2. Box, Jenkins (1976): Time Series Analysis – forecasting and Control, Holden Day

This covers both theory and practice very well but theory is advanced –very useful, if

not essential, for practising forecasters

3. Granger, Newbold (1986): Forecasting Economic Time Series, Academic Press A

very good account of the interaction of standard econometric methods and time series

methods. Some sections are difficult but much of the material will repay the effort

involved in mastering it

4. Jonathan D. Cryer and Kung-Sik Chan (2008), Time Series Analysis with

applications in R, Springer, 2nd edition. Gujarati D N. (2003). Basic

Econometrics, 4th edn. pp 465-7. McGraw-Hill Companies Inc., New York

5. Yule, G. U. (1926). “Why do we sometimes get nonsense-correlations between

time-series? A study in sampling and the nature of time-series.” Journal of the

Royal Statistical Society, 89, 1, 1–63.

6. Bussay, A. Velde, M., Fumagalli, D. and Seguini, L. 2015. Improving

perational maize yield forecasting in Hungary. Agricultural Systems, 141: 94-

106.

7. FICCI. 2018. Website: http://www.ficci.in/events-page. asp?evid=23706,

accessed on 10-07-2018.

45 | P a g e
8. Horie, T., Yajima, M. and Nakagawa, H. 1992. Yield forecasting. Agricultural

Systems, 40(1-3): 211-236.

9. Joshi, P.K., Singh, N.P., Singh, N.N., Gerpacio, R.V.and Pingali, P.L. 2005.

Maize in India: Production Systems, Constraints, and Research Priorities.

Mexico, D.F.: CIMMYT.

10. Manatsa, D., Nyakudya, I.W., Mukwada, G. and Matsikwa, H. 2011. Maize

yield forecasting for Zimbabwe farming sectors using satellite rainfall

estimates. Natural Hazards, 59(1): 447–463.

11. Martin, R.V., Washington, R. and Downing, T.E. 2000. Seasonal Maize

Forecasting for South Africa and Zimbabwe Derived from an

Agroclimatological Model. Journal of Applied Meteorology, 39: 1473-1479.

12. Santosha Rathod, Singh, K.N., Prawin Arya, Mrinmoy Ray, Anirban

Mukherjee, Kanchan Sinha, Prakash Kumar and Ravindra Singh Shekhawat.

2017. Forecasting maize yield using ARIMA-Genetic Algorithm approach.

Outlook on Agriculture, 46(4): 265-271.

13. Tahir, A. and Habib, N. 2013. Forecasting of maize area and production in

Pakistan. Crop Production, 2(2): 44-48.

46 | P a g e