analysis and prediction of nse indices (1)
Post on 13-Apr-2017
28 Views
Preview:
TRANSCRIPT
1
Analysis and Prediction of National Stock
Exchange Indices
A PROJECT REPORT SUBMITTED TO
DEPARTMENT OF STATISTICS, SHIVAJI UNIVERSITY,
KOLHAPUR
FOR THE DEGREE OF
MASTER OF SCIENCE
IN STATISTICS
BY
SAGARE AMRUT SUNIL DEPARTMENT OF STATISTICS,
SHIVAJI UNIVRESITY, KOLHAPUR-416004
2015-2016
2
CIRTIFICATE
This is to certify that the project report entitled “Analysis
and prediction of Nation Stock Exchange indices ”, being
submitted by Mr.Sagare Amrut Sunil, as a partial fulfillment for the
award of degree of M.Sc. in Statistics of Shivaji university,
Kolhapur, is a record of bonafide work carried out by her under
my supervision and guidance.
To the best of my knowledge the matter presented
in the project in the project has not been submitted earlier.
Place: Kolhapur
Date:
Mr. S. D.Pawar.
Project Guide
Dr. D. N. Kashid.
Head of department,
Department of
statistics,
Shivaji university,
Kolhapur.
3
Acknowledgement
We wish to thank Department of Statistics (Shivaji University,
Kolhapur) for giving us an opportunity to do a project.
This report has been prepared under the guidance of
Mr.S.D.Pawar. We would like to express our profound gratitude towards
him for her guidance we constructive throughout this project.
Also we would like to thank Dr. H. V. Kulkarni for their support,
suggestions and guidance for this project.
Finally we would like to thank HOD, teachers, non-teaching staff
and friends for their valuable co-operation in this project.
4
INDEX
Sr.no Contents Page no
1 Introduction and description of the
problem 6
2 Data collection 7
3 Objectives 8
4 Terminology 9
5 Methodology 12
6 Exploratory data analysis 13
7 Forecasting 20
8 Major finding 45
9 Limitations 46
10 References 47
5
1. Introduction and description of the problem:
The National Stock Exchange of India Limited (NSE) is the
leading stock exchange of India, located in Mumbai. NSE has a market
capitalization of more than US$1.65 trillion, making it the world’s 12th-
largest stock exchange as of 23 January 2015. NSE offers trading,
clearing and settlement services in equity, equity derivatives, debt and
currency derivatives segments. It is the first exchange in India to
introduce electronic trading facility thus connecting together the investor
base of the entire country. Trading on the equities segment takes place
on all days of the week (except Saturdays and Sundays and holidays
declared by the Exchange in advance).Time of trading is 9.00 am to 3.30
pm.
If we invest in bank we will get fix percentage of return on fix
deposit or on saving account which is dependent on bank, but in share
market we will get more return if we know the idea about share market
and where to invest the money. Sometimes price of share will go up or
down which is dependent on lots of factor. Overall there is risk to invest
money in share market as compare to bank. There are approximately
5000 companies are listed in NSE. Then main problem is that which
company we take to invest our money. Thanks to NSE there are lots of
indices which shows index of particular sector (like bank, IT, Energy,
Pharmacy and more) using this indices we make our strategy to invest.
Each sector contain number of companies in which we invest In this
project we will study on some sector indices of NSE to predict its future
value and to find confidence interval to minimize the risk to invest
money in share market using statistics. Using this we can make our
strategy to invest the money in which company of particular sector
whose index value shows positive response to make some money by
investment to become reach.
In this project we use daily closing price of Nifty50, Nifty next 50,
Bank, IT, FMCG, Pharma, Metal, Energy indices which can cover lots
of company in NSE. Using this value we try to fit Time series model to
predict future value to give some idea to invest.
6
2. Data collection: Data is collected from the official website of
National Stock Exchange of India Limited (NSE)
http://www.nseindia.com/products/content/equities/indices/historical_in
dex_data.htm
I collected data from the above website of following eight indices of
National Stock Exchange (NSE)
Nifty50
Next50
Bank
Energy
Metal
IT
FMCG
Pharma
This is in the form of Excel file of .csv format, which contain following
7 variables
Date
Open
High
Low
Close
Shares Traded
Turnover (Rs. Cr)
This is categorical data, in which open, high, low, close shows value of
index while share traded shows total number of share traded and
turnover shows turnover in ₨ on that perticular day
7
3. Objectives:
Following are our main objective of project,
To analysis the data of eight indices to find out some interesting
pattern from that to give idea about the market behavior.
To minimize the risk in share market by predicting the future index
value using time series analysis.
To find the confidence interval for forcasted value
8
4. Terminology:
1) Indices: An Index is used to give information about the price
movements of products in the financial, commodities or any other
markets. Financial indexes are constructed to measure price movements
of stocks, bonds, T-bills and other forms of investments. Stock market
indexes are meant to capture the overall behavior of equity markets. A
stock market index is created by selecting a group of stocks that are
representative of the whole market or a specified sector or segment of
the market. An Index is calculated with reference to a base period and a
base index value.
The different indices in stock market out of these we will study on 8
indices whose information is given below
Nifty 50 :
The Nifty 50 is a well diversified 50 stock index accounting for 13
sectors of the economy. It is used for a variety of purposes such as
benchmarking fund portfolios, index based derivatives and index funds.
Nifty Next 50
The Nifty Next 50 Index represents 50 companies from Nifty 100
after excluding the Nifty 50 companies.
FMCG
FMCGs (Fast Moving Consumer Goods) are those goods and
products, which are non-durable, mass consumption products and
available off the shelf. The Nifty FMCG Index comprises of maximum
of 15 companies who manufacture such products which are listed on the
National Stock Exchange (NSE).
9
Nifty Bank
Nifty Bank Index is an index comprised of the most liquid and
large capitalized Indian Banking stocks. It provides investors and market
intermediaries with a benchmark that captures the capital market
performance of Indian Banks. Index has 12 stocks from the banking
sector which trade on the National Stock Exchange.
Nifty IT
Companies in this index are those that have more than 50% of their
turnover from IT related activities like IT Infrastructure , IT Education
and Software Training , Telecommunication Services and Networking
Infrastructure, Software Development, Hardware Manufacturer’s,
Vending, Support and Maintenance.
Nifty Energy
Energy sector is universally recognized as one of the most
significant inputs for economic growth. Nifty Energy Index will include
companies belonging to Petroleum, Gas and Power sub sectors.
Nifty Metal
The Nifty Metal Index is designed to reflect the behavior and
performance of the Metals sector including mining. The Nifty Metal
Index comprises of maximum of 15 stocks that are listed on the National
Stock Exchange.
10
2) Share Trading
In our data we have daily share trade for each index which shows
total share traded by investor in companies which include in that index
that is number of buy and sell of shares by investors.
3) Closing Price
Closing value is the value of that index which shows daily closing
price or value of that index when market was close at that day
4) Turnover (Rs. Cr)
This is the total turnover in Rs in that index for one day by investors.
11
5. Methodology:
Collected data of indices are time dependent and I want to predict future
value of that indices. I used time series analysis to forcasting,
For this I used the following softwares:
R software
MINITAB
R package ‘forecast’ has been used to fit appropriate time series model
and to forecast the value. It contains the functions auto.arima( ) and
forecast( ) to fit time series model and forecasting.
R code:
w=read.csv(file.choose(),header=T) # To select Excel file in csv format
attach(w)
library(forecast)
fit=auto.arima(Close)
summary(fit)
f=forecast(fit,h=5)
plot(f)
f
12
6. Exploratory Data Analysis:
1) Closing Price
I ) Average of Indices –
The following table and graph represent the average of index of
particular index for each year
Average
Year Nifty 50 Bank Energy FMCG IT Next 50 Metal Pharma
2005 2268.91 3862.00 3898.77 3404.87 3066.22 4699.62 1100.14 2033.11
2006 3357.09 4735.47 5257.85 5273.30 4337.71 6143.26 1643.96 2555.71
2007 4571.29 6887.50 7462.06 5214.25 5054.30 8632.13 3140.22 2765.14
2008 4339.11 6564.79 8104.08 5540.57 3633.18 7378.32 3240.33 2928.15
2009 4113.96 6771.39 7985.37 6057.47 3738.89 7258.93 2902.21 2767.17
2010 5461.12 10359.09 9340.96 8219.75 6203.26 11494.92 4428.92 4241.57
2011 5335.91 10297.86 8395.28 9700.00 6375.92 10528.22 3663.80 4634.39
2012 5343.77 10509.24 7638.48 12632.03 6124.36 10363.78 2838.02 5277.25
2013 5915.90 11414.96 7872.61 16520.62 7488.17 11820.39 2247.90 6768.61
2014 7360.30 14522.65 8952.19 18387.19 10197.92 15447.97 2801.41 9022.92
2015 8285.91 18095.70 8377.88 20282.24 11585.63 19729.69 2155.71 12252.13
Average 5118.78 9449.10 7565.06 10102.81 6160.29 10311.12 2740.72 5016.82
From above line chart we visualias that,
FMCG is rapidly increases seems look like straight line
Metal shows constat behavior in last 10 year
0
5000
10000
15000
20000
25000
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Ind
ex
Val
ue
Average of Index per year
Nifty 50
Bank
Energy
FMCG
IT
Next 50
13
II) Maximum of Indices –
The following table and graph represent the maximum of index of
particular index for each year
Maximum
Year Nifty50 Bank Energy FMCG IT Metal Next50 Pharma
2005 2842.6 4692.85 4800.35 4421.43 3925.55 1273 5564.15 2272.45
2006 4015.95 6330.75 6134.2 6275.28 5432.25 2352.65 7213.9 2967.56
2007 6159.3 10090.7 11322.97 6329.6 5830.55 5456.99 12488.25 3173.55
2008 6287.85 10698.35 12012.26 6778.92 4748.2 5493.95 13069.45 3519
2009 5201.05 9526.7 9483.23 7526.72 5829.7 4709.33 10382.7 3850.16
2010 6312.45 13268.7 10195.42 9674.37 7503.65 5017.33 13555.15 5085.38
2011 6157.6 11894.75 9891.45 10762.35 7545.95 4800.65 12261 5172.9
2012 5930.9 12510.25 8318.3 15795.7 6747.2 3429.35 12340.05 6084.65
2013 6363.9 13317.1 8798 19407.85 9579.1 2986.05 12933.25 7731.55
2014 8588.25 18782.85 10603.35 21375.1 11981 3521.85 18970.25 11387.95
2015 8996.25 20555.25 9055.4 22295.15 12855.9 2715.45 21594.05 13831.15
Maximum 8996.25 20555.25 12012.26 22295.15 12855.9 5493.95 21594.05 13831.15
From above line chart we visualias that,
Bank and Next50 shows same behavior
FMCG shows rapid increase in between 2011to 2015
0
5000
10000
15000
20000
25000
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Ind
ex
Val
ue
Maximum of Index per year
Nifty 50
Bank
Energy
FMCG
IT
Metal
Next 50
Pharma
14
III) Minimum of Indices –
The following table and graph represent the maximum of index of
particular index for each year
Minimum
Year Nifty 50 Bank Energy FMCG IT Metal Next 50 Pharma
2005 1902.5 3084.6 3351.61 2679.45 2514.1 940.32 4019.65 1765.76
2006 2632.8 3428.15 4181.35 4251.16 3219.35 1166.42 4517.8 2058.4
2007 3576.5 4893.45 5429.74 4376.84 4172.75 1984.79 6295.4 2442.99
2008 2524.2 4053.2 4748.84 4362.96 2126.1 1188.78 3782.1 2126.66
2009 2573.15 3339.7 5364.89 4550.77 2002 1278.37 3595.1 1968.69
2010 4718.65 8223.25 8615.19 6885.71 5449.75 3806.32 9801.65 3481.72
2011 4544.2 7798.55 6968.1 8157.45 5087.65 2464.6 8295.85 4300.25
2012 4636.75 7995.05 6875.8 10103.95 5489.6 2495.1 8312.1 4567.6
2013 5285 8664.2 7058.25 14516.2 5972.7 1628.2 10203.1 5778.2
2014 6000.9 10102.1 7405.95 16336.1 8675.1 2142.55 11729.65 7411.6
2015 7558.8 15790.1 7299.15 18828.45 10798.25 1605.45 18308.85 10604.95
Minimum 1902.5 3084.6 3351.61 2679.45 2002 940.32 3595.1 1765.76
From above line chart we visualias that,
FMCG shows maximum increase in minimum index.
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
20000
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Ind
ex
Val
ue
Minimumof Index per year
Nifty 50
Bank
Energy
FMCG
IT
Metal
Next 50
Pharma
15
Return from Indices –
Following table and graph shows the return in percentage from each
index for one year. The red colure digit shows negative return and green
colure digit shows positive return.
Return = 𝐸𝑛𝑑𝑖𝑛𝑔 𝑝𝑟𝑖𝑐𝑒−𝑠𝑡𝑎𝑡𝑟𝑡𝑖𝑛𝑔 𝑝𝑟𝑖𝑐𝑒
𝑠𝑡𝑎𝑡𝑟𝑡𝑖𝑛𝑔 𝑝𝑟𝑖𝑐𝑒 × 100
Year Nifty 50 Bank Energy FMCG IT Metal Next 50 Pharma
2006 39.861422 31.87929 19.55762 17.86001 39.85325 92.32865 27.967406 25.62271 2007 53.181614 63.31432 95.42174 21.80459 -12.8136 136.3008 74.206441 14.08794 2008 -51.83949 -49.5061 -48.3737 -21.4194 -53.9404 -73.6483 -63.95465 -25.42643 2009 71.456592 76.47634 58.13267 41.00342 155.1762 210.1549 122.32286 58.343024 2010 17.245136 29.40076 3.545622 29.01356 27.45927 -1.72968 16.407023 35.456616 2011 -24.90094 -32.7866 -28.7365 8.410411 -18.0642 -48.6611 -32.03572 -10.41523 2012 27.354289 56.02467 12.65232 50.19126 -3.10002 16.23783 48.458873 32.126281 2013 5.9344463 -10.0275 0.362435 11.56614 57.75529 -15.9748 3.5384769 26.060001 2014 31.437005 64.56445 8.941337 18.03911 18.53046 6.706632 43.81788 42.811681 2015 -4.07593 -9.75043 -0.54281 0.673311 -0.02809 -32.1693 6.630424 9.5232646
Over all 180.20064 271.4063 78.31792 359.7315 188.6671 54.08914 259.73619 442.59189
From above we visualise that. In year 2008 due to
Economical wide recetion in all over the world affect share market badly
but after that in 2009 share market give outstanding return in some index
such as IT, Metal and Next50 as compare to other.
-100
-50
0
50
100
150
200
250
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015Ret
urn
in p
erce
nta
ge
Return from each Index per year
Nifty 50
Bank
Energy
FMCG
IT
Metal
Next 50
16
2) Share Trading
I) Average Trading-
Above graph shows that, the investors interested in trading in the
Nifty50 as compare to others.
II) Total Trading-
Below graph represent total share traded in per year.
This shows that, Investors are interested in company which is in Nifty50
and Next50 to trading.
0
20000000
40000000
60000000
80000000
100000000
120000000
140000000
160000000
180000000
Total
Average Share Trading In Last 10 Year
Bank
Energy
IT
Metal
Next50
FMCG
Nifty50
Pharma
0
2E+10
4E+10
6E+10
8E+10
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Total Share Trade
Bank
Energy
IT
Metal
Next50
FMCG
Nifty50
17
3) Turnover (Rs.Cr)
I) Average Turnover-
Below graph give us visualisation for average turnover in indices.
II) Maximum Turnover-
Above bar plot give visualisation that, turnover is maximum in Nifty50
index which is 32873.45. (Cr)
0
1000
2000
3000
4000
5000
6000
7000
Total
Average Turnover In Last 10 Year
Bank
Energy
IT
Metal
Next50
FMCG
Nifty50
Pharma
0
5000
10000
15000
20000
25000
30000
35000
Total
MaximumTurnover In Last 10 Year
Bank
Metal
Energy
IT
Next50
FMCG
Nifty50
Pharma
18
II) Minimum Turnover-
Above bar plot visualise that, second maximum turnover pharma here
has lower turnover which is 1.75 (Cr.) among all indices.
III) Total Turnover-
Below graph represent total turnover in per year.
From above graph we conclude that, investors favorite index is Nifty50
to invest because in 10 years total turnover in Nifty50 is more than
others.
0
20
40
60
80
100
120
Total
MinimumTurnover In Last 10 Year
Bank
Metal
Energy
IT
Next50
FMCG
Nifty50
Pharma
0
500000
1000000
1500000
2000000
2500000
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
Total Turnover (Rs. Cr)
Bank
Energy
IT
Metal
Next50
FMCG
Nifty50
19
7. Forecasting:
Now in this section we fit the time series model to closing value of each
index. For this analysis we take only one year data (i.e 2015) to
prediction the future value and to find 95% confidence interval.
Nifty Bank
Time series plot of closing value:
Here we see that decreasing trend. We take difference to remove
trend.
Time series plot of difference:
20
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
21
Using R-code I get the following prediction
Output:
ARIMA (0, 1, 0) with drift
Coefficients:
drift
-7.4018
s.e. 16.3120
sigma^2 estimated as 65990: log likelihood=-1720.49
AIC=3444.97 AICc=3445.02 BIC=3451.99
Forecast:
Forecast Lo 95 hi 95 Actual
16914.8 16411.31 17418.29 17039.25
16907.4 16195.36 17619.44 16599.15
16899.99 16027.93 17772.06 16542.5
16892.59 15885.62 17899.57 16433.15
16885.19 15759.36 18011.02 16073.85
Graph:
14500
15000
15500
16000
16500
17000
17500
18000
18500
1 2 3 4 5
Ind
ex
valu
e
Forecast, CI & Actual of Nifty Bank
Forecast
Lo 95
hi 95
Actual
22
Nifty Bank
Time series plot of closing value:
Here we see that decreasing trend. We take difference to remove
trend.
Time series plot of difference:
23
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
24
Using R-code I get the following prediction
Output:
ARIMA (0, 1, 0) with drift
Coefficients:
drift
0.5468
s.e. 14.8837
sigma^2 estimated as 54941: log likelihood=-1697.85
AIC=3399.71 AICc=3399.76 BIC=3406.73
Forecast:
Forecast Lo 95 hi 95 Actual
20193.2 19733.79 20652.6 20184.95
20193.74 19544.05 20843.44 20087.8
20194.29 19398.58 20990 20047.45
20194.84 19276.03 21113.65 19689.45
20195.38 19168.12 21222.64 19261.55
Graph:
18000
18500
19000
19500
20000
20500
21000
21500
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty FMCG
Forecast
Lo 95
hi 95
Actual
25
Nifty IT:
Time series plot of Closing value:
Here we see that decreasing trend. We take difference to remove
trend.
Time series plot of difference:
26
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
27
Using R-code I get the following prediction
Output:
ARIMA(0,1,0)
sigma^2 estimated as 16136: log likelihood=-1547.04
AIC=3096.09 AICc=3096.1 BIC=3099.6
Forecast:
Forecast Lo 95 hi 95 Actual
11212.55 10963.58 11461.52 11174.85
11212.55 10860.45 11564.65 11029.25
11212.55 10781.32 11643.78 10997.15
11212.55 10714.61 11710.49 11018.15
11212.55 10655.83 11769.27 10863.2
Graph:
10000
10200
10400
10600
10800
11000
11200
11400
11600
11800
12000
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty IT
Forecast
Lo 95
hi 95
Actual
28
Nifty Metal:
Time series plot of closing value:
Here we see that decreasing trend. We take difference to remove
trend.
Time series plot of difference:
29
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
30
Using R-code I get the following prediction
Output:
ARIMA(0,1,0) with drift
Coefficients:
drift
-3.5077
s.e. 2.1576
sigma^2 estimated as 1155: log likelihood=-1220.83
AIC=2445.66 AICc=2445.71 BIC=2452.68
Forecast:
Forecast Lo 95 hi 95 Actual
1823.342 1756.745 1889.94 1830.3
1819.835 1725.651 1914.018 1804.9
1816.327 1700.976 1931.678 1862.7
1812.819 1679.624 1946.015 1829.05
1809.312 1660.395 1958.229 1742.55
Graph:
1500
1550
1600
1650
1700
1750
1800
1850
1900
1950
2000
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty Metal
Forecast
Lo 95
hi 95
Actual
31
Nifty Next50:
Time series plot of closing value:
Here we see that normal increasing trend. We take difference to
remove trend.
Time series plot of difference:
32
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
33
Using R-code I get the following prediction
Output:
ARIMA(0,1,0) with drift
Coefficients:
drift
5.0291
s.e. 14.4869
sigma^2 estimated as 52050: log likelihood=-1691.18
AIC=3386.36 AICc=3386.41 BIC=3393.38
Forecast:
Forecast Lo 95 hi 95 Actual
19982.08 19534.92 20429.24 20169.45
19987.11 19354.73 20619.48 19956.85
19992.14 19217.64 20766.64 20068.75
19997.17 19102.85 20891.48 20008.7
20002.2 19002.32 21002.07 19408.15
Graph:
18000
18500
19000
19500
20000
20500
21000
21500
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty Next50
Forecast
Lo 95
hi 95
Actual
34
Nifty Pharma:
Time series plot of closing value:
Here we see that slightly increasing trend. We take difference to
remove trend.
Time series plot of difference:
35
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
36
Using R-code I get the following prediction
Output:
ARIMA(0,1,0) with drift
Coefficients:
drift
4.2115
s.e. 10.9844
sigma^2 estimated as 29924: log likelihood=-1622.82
AIC=3249.63 AICc=3249.68 BIC=3256.65
Forecast:
Forecast Lo 95 hi 95 Actual
11967.71 11628.67 12306.76 11979.85
11971.92 11492.44 12451.4 11733.7
11976.13 11388.89 12563.38 11741.5
11980.35 11302.26 12658.44 11673.8
11984.56 11226.43 12742.68 11451.65
Graph:
Nifty Energy:
10000
10500
11000
11500
12000
12500
13000
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty Pharma
Forecast
Lo 95
hi 95
Actual
37
Time series plot of closing value:
There is one problem has occurred in time series plot. we observe from
above time series plot after some 150 day there market is suddenly
down. From data we observed that on 24 Aug 2015 market was suddenly
down by 664.75 point. If we go through one year data to analyze the
data this variation will affect on analysis.
So we take data from 25th Aug 2015 to 31st Dec 2015 to analysis data as
follows
38
Time series plot of closing value:
Here we see that normal increasing trend. We take difference to
remove trend.
Time series plot of difference:
39
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
40
Using R-code I get the following prediction
Output:
Series: Close
ARIMA(0,1,0)
sigma^2 estimated as 8622: log likelihood=-505.75
AIC=1013.49 AICc=1013.54 BIC=1015.93
Forecast:
Forecast Lo 95 hi 95 Actual
8584.1 8402.106 8766.094 8594.7
8584.1 8326.722 8841.478 8468.6
8584.1 8268.877 8899.323 8555
8584.1 8220.112 8948.088 8662.45
8584.1 8177.149 8991.051 8454.45
Graph:
7600
7800
8000
8200
8400
8600
8800
9000
9200
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty Energy
Forecast
Lo 95
hi 95
Actual
41
Nifty Nifty50:
Time series plot of closing value:
Here we see that decreasing trend. We take difference to remove
trend.
Time series plot of difference:
42
ACF and PACF of difference:
For ARIMA (p, d, q) model here ARIMA (0, 1, 0)
Normality graph for difference is:
43
Using R-code I get the following prediction
Output:
ARIMA(0,1,0) with drift
Coefficients:
drift
-1.3670
s.e. 5.3621
sigma^2 estimated as 7131: log likelihood=-1445.69
AIC=2895.37 AICc=2895.42 BIC=2902.39
Forecast:
Forecast Lo 95 hi 95 Actual
7944.983 7779.476 8110.49 7963.2
7943.616 7709.553 8177.679 7791.3
7942.249 7655.582 8228.916 7784.65
7940.882 7609.867 8271.896 7741
7939.515 7569.43 8309.6 7568.3
Graph:
7000
7200
7400
7600
7800
8000
8200
8400
1 2 3 4 5
Ind
ex
Val
ue
Forecast, CI 95% & Actual of Nifty Nifty50
Forecast
Lo 95
hi 95
Actual
44
8. Major Finding:
Investors are interested in Nifty50 and Next50 index sector.
FMCG index shows more return than any other in last decade
Decreasing in index gives opportunity to investors to invest his
money to make profit from decreased share if, investor know when
market will go up.
From return table observe that the return given by all index in 10
years are positive excpet some years in between these year. This
shows that investor should invest his money for long term to get
better result for make some money.
Using time series forecasting we find the future behavior of index
value and can give a 95% confidence interval for that index.
From above time series analysis I found that for all eight index the
time series model is same that is ARIMA (0, 1, 0).
Time series analysis shows that the market will down for few days.
45
9. Limitations:
To forecast the future value of index is calculated using only
closing price of index. Analysis does not considore other factores
which affect on the share market that’s why the pridicted value
does noat give you exact future value. There are lots of factors
affecting on share price such as, natural disaster, government
decisions, terror attack, International level activity, global
economical recetion etc.
These time series analyses give you only the future behaviour of
market not exact prediction.
46
10. Reference
Peter J. Brockwell, Richerd A. Davis(1987), Introduction to time
series and forecast.
Tsay R. S. Analysis of financial time series
47
top related