Academy of Strategic Management Journal (Print ISSN: 1544-1458; Online ISSN: 1939-6104)

Research Article: 2018 Vol: 17 Issue: 2

Predicting Stock Market Movements of India using Data Deterministic Approach

Marxia Oli Sigo, Bharathidasan University

 

Murugesan Selvam, Bharathidasan University

Sankaran Venkateswar, Trinity University

Sankaran Venkateswar, Trinity University

Sankarkumar Amirdhavasani, Bharathidasan University

Chinnadurai Kathiravan, Bharathidasan University

Keywords

Behavioural Finance, Capital Market, Predictive Analytics, Stochastic, Stock Index.

JEL Codes

C45, C53, E27, E44, G1

Introduction

Forecasting the stock price movement of a company and stock index is a classic problem. Efficient Market Hypothesis clearly asserts that it is not possible to exactly predict the stock prices of corporate entities, due to the existence of random walk behaviour in stock markets (Fama, 1970). The movements of stock prices and stock indices are influenced by many macro-economic variables such as political events, policies of the corporate enterprises, general economic conditions, commodity price index, bank rate, loan rates, foreign exchange rates, investors’ expectations, investors’ choices and the human psychology of stock market investors (Miao et al., 2007). Neural networks are a class of generalized, non-linear and non-parametric models, inspired by the studies of human brain. The feed-forward networks are the most widely used (Ou & Wang, 2009). Prediction of stock market movements has become increasingly difficult due to uncertainties, involved with the probable future outcomes. At a particular point of time, there could be trends, cycles and random walk or a combination of three cases/events (Robert & David, 2011). Closing price of a stock/index has been used, as one of the important statistical data, to derive useful information about the current and probable future movement pattern of stock market (Defu et al., 2005). In data deterministic approach, a layer is employed to convert each of the technical indicator’s continuous value, from +1 to -1, indicating the probable future growth/decline movements. This layer explains the manner of stock market movements, on both upwards and downwards direction, across the time periods (Shuai & Wei, 2014). The data deterministic approach could forecast the future trend of stock market and to provide stock information signs, for taking better investment decision of buying and selling of stocks by the investors (Jigar et al., 2015a).

Review Of Literature

An extensive review of literature, in the area of prediction of stock indices, has been done by the researchers. Prediction of stock price movements of individual stocks and stock market indices is explained by Amitai (1976), who also explained the difficulties involved in making specific and accurate forecasting of financial markets. Wang & Leu (1996) forecasted six week stock price trend, based on past four years stock price movements of Taiwan stock market, by using recurrent neural network. Hansen & Nelson (2003) applied a time-delay neural network to predict the stock price movements and the results of future trend prediction, using the hybrid system, proved to be promising. Kim (2003) used twelve technical indicators, to make prediction of daily stock price changes and stock index movements of Korea Composite Stock Price Index (KOSPI). Simulation results of Shanghai Composite Index show that neural networks could be applied to maximize the returns of stock market investment (Defu et al., 2005). Franklin Kuo classified the networks into linear, passive, reciprocal, causal and time invariant and each network approaches has unique and different inherent characteristic properties accordingly. Teo & Douglas (2006) calculated the profitability of stock indices, based on daily trades of S&P 500 (India), DAX (Germany), Topix (Japan) and FTSE (UK). Andre & Beatriz (2008) adopted neural network, to forecast the stock market returns of emerging economies. Ou & Wang (2009) tried out ten different data mining techniques, in order to predict the stock price movements of Hang Seng index of Hong Kong stock market. According to Robin & Ryan (2009), novelty, complexity and anonymity of forecasting the stock markets, influenced the observers to have reservations about the prediction outcome of stock price movements. Nair et al. (2011) predicted the closing value of next day, for five international stock indices, using an adaptive artificial neural network system. Garg et al. (2013) analysed the model selection criteria, for two data transmission models of the stocks, indexed in New York Stock Exchange. Jigar et al. (2015a) used regression method to predict the trends. Prediction performance of BSE-Sensex, NSE-Nifty, Reliance Industries and Infosys Limited were measured and compared in a group of stock indices and individual stocks respectively. Shin & Shie (2015) used the Box-Jenkins model, spectral analysis and Markov process, to forecast the stock prices. Based on the above reviews, the researchers applied one of the neural network methods i.e., data deterministic approach is used in this study, to predict the robust growing Indian stock market.

Statement Of The Problem

Prediction of the movements of financial markets is one of the classic issues. Profit making in stock market investments is, linked to the level of financial literacy, financial intelligence and financial investment practice. Less exposure to these aspects, among the financial investors and absence of proven forecasting techniques, to exactly predict the probable futuristic movement pattern of stock price/index values, cause the magnitude and severity of this issue which also vindicates this kind of study (Melek & Derya, 2010).

Need Of The Study

This study would help the investors, ranging from domestic retail investors, financial institutions, mutual funds, investment banks, to foreign institutional investors, to take timely and well-informed investment decisions, based on scientific thinking and rational approach (Amitai, 1976). Availability of alternate investment options, absence of prudent prediction methods and incidence of lower level of financial literacy reiterate this kind of study.

Objective

The primary objective is to find out the existing stock index movement pattern and to predict the probable future movements of BSE-Sensex and NSE-Nifty.

Hypotheses

NH1: There is no corresponding relationship between the upward/downward movements of BSE-Sensex and NSE-Nifty, during the pre-global financial crisis period.

NH2: There is no corresponding relationship between the upward/downward movements of BSE-Sensex and NSE-Nifty, during the post-global financial crisis period.

Research Methodology

Sampling Design of the Study

Based on the free-float market capitalization, as on 07/02/2017, S&P BSE-Sensex (Rs. 53,88,277 Crores) and CNX NSE-Nifty (Rs. 49,27,183 Crores) were considered, since these two stock indices signify the overall direction of the stock market movements in India.

Sources of Data

The secondary data of the daily closing stock index values of BSE-Sensex and NSE-Nifty were collected from respective websites of Bombay Stock Exchange (www.bseindia.com) and the National Stock Exchange of India (www.nseindia.com).

Period of the Study

The study focused on the behaviour of stock price/index movements of eighteen years from 1st January 1999 to 31st December 2016, before and after the global financial crisis.

Statistical Tools Used in the Study

To forecast the movements of BSE-Sensex and NSE-Nifty, ten technical indicators (Simple Moving Average, Weighted Moving Average, Momentum, Stochastic K%, Stochastic D%, Relative Strength Index, Moving Average Convergence and Divergence, Larry Williams R, A/D Oscillator and Commodity Channel Index) were used.

Limitations of the Study

Each technical indicator has its own optimization about the stock price movements. Only two stock indices, BSE-Sensex and NSE-Nifty, were used as sample.

Data Analysis And Interpretation

The daily closing values of stock indices of BSE-Sensex and NSE-Nifty, which were collected and analysed for this study, were as follows:

The Observations of the Total Dataset

Observation of BSE-Sensex

Table 1 exhibits the observations of the total dataset (continuous), relating to BSE-Sensex, during the study period from 1999 to 2016. The analysis of pre-crisis period (i.e., 1999 to 2007) shows that in 1999, the number of bull movements was recorded as 136, which represented 54.83% while there were 112 bear movements, which accounted for 45.17%, out of total stock index movements of 248. In the next year (i.e., 2000), the bear movements of 128 (51.20%) outnumbered the occurrence of bull movements of 122 (48.80%) and this symbolized downward trend of the stock index (BSE-Sensex) movements, due to the dotcom bust/bubble in the stock markets of United States of America (USA), caused by the Information Technology Sector, which led to the crash of stock markets all over the Globe.

Table 1: Observations Of The Dataset (Continuous) For Bse-Sensex
Year Stock Index Movements
Increase (+) Decrease (-) Total
Pre-Global Financial Crisis Period
1999 136 (54.83%) 112 (45.17%) 248
2000 122 (48.80%) 128 (51.20%) 250
2001 134 (54.03%) 114 (45.97%) 248
2002 141 (56.17%) 110 (43.83%) 251
2003 146 (57.48%) 108 (42.52%) 254
2004 136 (53.54%) 118 (46.46%) 254
2005 147(58.56%) 104 (41.44%) 251
2006 148 (59.20%) 102 (40.80%) 250
2007 137 (55.46%) 110 (44.54%) 247
Total (A) 1247 (55.34%) 1006 (44.66%) 2253
Post-Global Financial Crisis Period
2008 101(41.05%) 145 (58.95%) 246
2009 116 (47.73%) 127 (52.27%) 243
2010 134 (53.17%) 118 (46.83%) 252
2011 127 (51.41%) 120 (48.59%) 247
2012 128 (50.99%) 123 (49.01%) 251
2013 121 (48.40%) 129 (51.60%) 250
2014 126 (51.63%) 118 (48.36%) 244
2015 129 (52.01%) 119 (47.99%) 248
2016 126 (51.01%) 121 (48.99%) 247
Total (B) 1108 (49.73%) 1120 (50.27%) 2228
Sum=A+B 2356 (52.57%) 2125 (47.43%) 4481

In the mid pre-crisis period (during 2003), of the stock index upward movements were recorded as 146 (57.48%) whereas the index downward movements of BSE-Sensex were recorded as 108 (42.52%). Out of 2253 (total stock index movements), during the pre-crisis period of nine years (1999-2007), a high number of upward stock index movements were registered as 1247 (55.34%), than that of downward stock index movements of 1006 (44.66%).

In the year of Global Financial Crisis (i.e., 2008), the downward movements of 145, at 58.95%, outnumbered the upward movements of 101, at 41.05%. This clearly indicated, due to the global financial crisis, which highly shattered the stock markets worldwide. Similar trend was registered in 2009, that the bear movements of 127 (52.27%) were more than the bull movements of 116 (47.73%). In the mid post-crisis period (2012), the upward movements of stock index were recorded as 128 (50.99%), while the downward movements were recorded as 123 (49.01%). In 2016, the upward movements of 126 (51.01%) were marginally greater than downward movements of 121 (48.99%).

The analysis of post-crisis period of nine years (2008-2016) brought out the fact that the downward stock index movements of 1120 (50.27%) were more than the upward stock index movements of 1108 (49.73%) and this clearly indicated the negative impact of the crisis.

According to Table 1, the total observations, for the entire study period of 18 years (1999-2016), were registered as 4481, out of which, 2356 incidents (52.57%) were recorded as upward movements while 2125 incidents (47.43%) were registered as downward movements. It is suggested that investors could observe and study these movements of stock markets so as to devise their investment strategies in such a way to earn higher returns in the stock market.

Observation of NSE-Nifty

The total observations of the dataset (continuous), relating to NSE-Nifty, during the study period from 1999 to 2016, are shown in Table 2. During pre-crisis period (1999 to 2007), there were total stock index movements of 248 and a high number of bull movements were recorded as 134 (54.03%) in 1999, than that of bear movements, which were 114 (45.87%) in 1999.

Table 2: Observations Of The Dataset (Continuous) For Nse-Nifty
Year Stock Index Movements
Increase (+) Decrease (-) Total
Pre-Global Financial Crisis Period
1999 134 (54.03%) 114 (45.87%) 248
2000 121 (48.40%) 129 (51.60%) 250
2001 135 (54.43%) 113 (45.57%) 248
2002 141 (56.17%) 110 (43.83%) 251
2003 146 (57.48%) 108 (42.52%) 254
2004 137 (53.93%) 117 (45.07%) 254
2005 145 (57.76%) 106 (42.24%) 251
2006 147 (58.80%) 103 (41.20%) 250
2007 125 (50.60%) 122 (49.40%) 247
Total (A) 1231 (54.63) 1022 (45.17%) 2253
Post-Global Financial Crisis Period
2008 102 (41.46%) 144 (58.54%) 246
2009 107 (44.03%) 136 (55.97%) 243
2010 130 (51.58%) 122 (48.42%) 252
2011 128 (51.82%) 119 (48.18%) 247
2012 129 (51.39%) 122 (48.61%) 251
2013 129 (51.60%) 121 (48.40%) 250
2014 126 (51.63%) 118 (48.37%) 244
2015 129 (52.01%) 119 (47.99%) 248
2016 125 (50.60%) 122 (49.40%) 247
Total (B) 1105 (49.59%) 1123 (50.41%) 2228
Sum=A+B 2336 (52.13%) 2145 (47.87%) 4481

The downward trend of daily closing value of the stock index movements, in 2000, was attributed to the bust/bubble of Information Technology Sector stocks in United State of America (USA), which led to the crash of stock markets all over the Globe. In 2000, the number of bear movements were129 (51.60%), which was slightly greater than the number of bull movements of 121 (48.80%). In the mid post-crisis period (2003), the upward movements of the stock index were 146 (57.48%) and the downward movements were108 (42.52%). At the end of 2007 (i.e., before the occurrence of Global Financial Crisis), the bull movements were at 125 (50.60%) while the bear movements were at 122 (49.40%). A high number of stock index movements were recorded upward as 1231 (54.63%) than the downward movements, at 1022 (45.17%), out of 2253 total stock index movements, during the pre-crisis period of 1999-2007.

In the year of Global Financial Crisis (i.e., 2008), the downward movements of 144 (58.54%) outnumbered the upward movements of 102 (41.46%). There was the sharpest and the widest downfall in NSE-Nifty movements in 2008, due to the financial crisis, which severely shattered the stock markets worldwide. It is interesting to note that the negative impact continued in the stock market during 2009 also, which signified the fact that the number of bear movements of 136 (55.97%) outnumbered the bull movements of 107 (44.03%) in 2009. In the mid pre-crisis period (during 2012), the upward movements of stock index NSE-Nifty were recorded as 129 (51.60%) while the downward movements of Nifty were registered as 122 (48.61%). The last year of study (2016) vividly indicated that the upward stock movements of 125, at 50.60% was relatively more than the downward stock movements of 122, at 49.40%.

According to Table 2, the downward stock index movements was recorded as 1123, at 50.41%, which was more than the upward stock index movements of 1105, at 49.59%, this signified the negative impact of the crisis. Out of the total observations of 4481, a higher number of observations of 2336 (52.13%) were recorded as upward movements while 2145 observations (47.87%) were recorded as downward movements, during the post-crisis period 2008-2016.

It is inferred from Tables 1 and 2 BSE-Sensex and NSE-Nifty were rampant and robust in terms of stock market movements. A holistic view of the entire study period (1999-2016) revealed that there was marginally a higher number of bullish movements (Sensex: 2356 and Nifty: 2336) than bearish movements (Sensex: 2145 and Nifty: 2125) out of the total movements of 4481. Individual investors are advised to take note of these stock index movements, in order to devise their investment strategies appropriately and earn better returns in the stock markets.

Parameter Setting Data and Comparison Dataset (Discrete)

Cases for BSE-Sensex

For the purpose of achieving the objective of the study of finding out the existing stock index movement pattern and predicting the probable future movements, 50% of each discrete dataset were used for the training purpose and hold-out dataset. The parameter setting and comparison dataset (discrete), relating to BSE-Sensex are depicted in Table 3. In 1999, 126 observations were used as training data, covering upward movements of 69 and downward movements of 57 whereas 122 instances were used as the hold-out dataset (upward movements of 66 and downward movements of 56), out of 248 total observations. In 2000, the observations of datasets (training and hold-out) were 127 and 123 respectively and the bearish movements (68-training and 66-hold-out) were more than the bullish movements (59-training and 57-hold-out). This was mainly due to the dotcom burst, i.e., stock prices of Information Technology Sector had fallen heavily, in the stock markets, throughout the world, including the USA.

Table 3: Parameter Setting And Comparison Dataset (Discrete) Of Bse-Sensex
Year Training Hold-out Total=A+B
Upward Downward Total(A) Upward Downward Total(B)
Pre-Global Financial Crisis Period
1999 69 57 126 66 56 122 248
2000 59 68 127 57 66 123 250
2001 72 57 129 65 54 119 248
2002 69 62 131 64 56 120 251
2003 70 60 130 65 59 124 254
2004 71 58 129 69 56 125 254
2005 69 59 128 66 57 123 251
2006 73 53 126 72 52 124 250
2007 63 62 125 62 60 122 247
Total (A) 615 536 1151 586 516 1102 2253
Post-Global Financial Crisis Period
2008 52 72 124 51 71 122 246
2009 53 70 123 52 68 120 243
2010 66 63 129 62 61 123 252
2011 65 61 126 61 60 121 247
2012 64 63 127 63 61 124 251
2013 67 61 128 62 60 122 250
2014 63 62 125 60 59 119 244
2015 65 63 128 62 58 120 248
2016 64 62 126 61 60 121 247
Total (B) 559 577 1136 534 558 1092 2228
Sum=A+B 1174 1113 2287 1120 1074 2194 4481

Source: www.bseindia.com (Using SPSS 20.0).

In 2003, the movements of growth and downfall, for datasets (training and hold-out), were at 70, 65 and 60, 59 respectively. In 2007, bull movements were 63 and 62 and bear movements were 62 and 60. The Indian stock market was optimistic since the total bull movements were (615 and 586) more than the bear movements (536 and 516). This indicated euphoric phenomenon in the stock market, during 1999-2007.

In 2008, bear movements (72 and 71) were more than the bull movements (52 and 51). Similar trend was continued in the next year (2009) also. In 2012, the upward movements (64 and 63) were higher than downward movements (63 and 61), on training and hold-out strategic dataset. In 2016, upward/downward movements, in case of training data, were at 64 and 62, 61 and 60 for the hold-out dataset. The Indian stock market was pessimistic since the total quantum of bear movements were (577 and 558) more than the bull movements (559 and 534), indicating the sober phenomenon in the post-crisis period. Investors have to note this trend to devise investment strategies in making investment in this stock index.

Cases for NSE-Nifty

The detailed case of the parameter setting and comparison dataset (discrete) of NSE-Nifty is provided in Table 4. In 1999, 126 were used as training data (upward: 68 and downward: 58) whereas 122 instances were used as the hold-out dataset (upward: 66 and downward: 56), out of 248 total observations (i.e., daily closing index value). In 2000, the observations were at 127 and 123 respectively while the bear movements were at 70 and 67, which were higher than the bull movements (57 and 56), due to the dotcom burst in the Information Technology Sector. In 2003, the growth and downfall movements were at 71 and 66 and 59and 58. In 2007 the bull movements were at 64 and 63 and bear movements were at 61 and 59.

Table 4: Parameter Setting And Comparison Dataset (Discrete) Of Nse-Nifty
Year Training Hold-out Total=A+B
Upward Downward Total(A) Upward Downward Total(B)
Pre-Global Financial Crisis Period
1999 68 58 126 66 56 122 248
2000 57 70 127 56 67 123 250
2001 71 58 129 65 54 119 248
2002 69 62 131 64 56 120 251
2003 71 59 130 66 58 124 254
2004 72 57 129 69 56 125 254
2005 70 58 128 66 57 123 251
2006 74 52 126 73 51 124 250
2007 64 61 125 63 59 122 247
Total (A) 616 535 1151 588 514 1102 2253
Post-Global Financial Crisis Period
2008 50 74 124 49 73 122 246
2009 53 70 123 52 68 120 243
2010 68 61 129 64 59 123 252
2011 64 62 126 62 59 121 247
2012 65 62 127 64 60 124 251
2013 66 62 128 62 60 122 250
2014 64 61 125 61 58 119 244
2015 65 63 128 61 59 120 248
2016 64 62 126 60 61 121 247
Total (B) 559 577 1136 535 557 1092 2228
Sum=A+B 1175 1112 2287 1123 1071 2194 4481

The NSE stock market was optimistic since the total bull movements were (616 and 588) more than the bear movements (535 and 514). This further indicated euphoric phenomenon, during1999-2007. In 2008, bear movements (74 and 73) was more than the bull movements (50; 49), due to the financial market crashing heavily throughout the world. Similar trend was continued in 2009 also. Overall, the Nifty was pessimistic since the total bear movements were (577 and 557) more than the bull movements (559 and 535), which revealed the sober trading in the post-crisis period (2008-2016).

It is inferred from the analysis of Tables 3 and 4 BSE-Sensex and NSE-Nifty were robust and rampant. Index movements (upward/downward or bull/bear or growth/decline) and the occurrence of such events was found to be relatively similar, throughout the study period. Investors should note this trend before devising their investment strategies in this stock index.

Data Validation using Technical Indicators

In order to predict future movements of stock indices, ten technical indicators were used. The first two technical indicators were Simple Moving Average (SMA) and Weighted Moving Average (WMA). Moving Average is a simple technical analysis tool signifies stock price data, by creating constantly updated average prices. If the stock price/index is above/below the value of moving average, then the stock price/index trend is upward or downward. There are three stochastic oscillators, namely Stochastic K%, Stochastic D% and Larry Williams R % are the clear trend indicators. Whenever the value of stochastic oscillator is increasing, the stock prices are likely to go up. On the contrary, when the value of stochastic oscillator is decreasing, the stock prices are likely to come down. If the value of a stochastic oscillator, at time ‘t’, is greater than the value, at time ‘t-1’, then the opinion of trend is termed as ‘up’ and represented as ‘+1’. If the value of stochastic oscillators, at time ‘t’, is less than the value at time ‘t-1’, then the opinion of trend is termed as ‘down’ and represented as ‘-1’.

Moving Average Convergence and Divergence (MACD) follows the trend of the stock. The Relative Strength Index (RSI) is generally used, to identify the overbought and oversold position/point. It ranges between 0 and 100. If the value of RSI exceeds 70 level, the stock is overbought and the price of the stock/index may fall in the near future (indicating opinion‘-1’). If the value of RSI goes below 30 level, the stock is oversold and the stock price/index may go up in the near future (indicating opinion‘+1’). Commodity Channel Index (CCI) measures the difference between change in the price of a stock/index and its average price/value changes. High positive readings indicate that prices are well above their averages, which is a show of strength. Low negative readings indicate the fact that prices are well below their averages, which is a show of weakness. It is used to identify overbought and oversold positions/levels of stock purchase/sale. The study has set +200 as overbought levels and -200 as oversold levels.

Accumulation/Distribution (A/D) Oscillator value also indicates the movement pattern of stock trend. If its value at time ‘t’ is greater than at time ‘t-1’, then the opinion of the stock index trend is ‘+1’ and vice versa. Momentum measures the rate of rise and fall of stock prices/index movements. Positive value of momentum indicates ‘up’ trend, whereas the negative value of momentum indicates ‘down’ trend (Jigar et al., 2015a).

Validation for BSE-Sensex

Table 5 exhibits the statistical values of ten technical parameters used to validate both training and hold-out Dataset, in respect of BSE-Sensex, during 1999 to 2007. The Simple Moving Average (SMA) earned a maximum value of 20375.87, minimum value of 2600.12, mean value of 6588.31 and standard deviation value of 4120.63. The Weighted Moving Average (WMA) attained the maximum value of 20142.87 and minimum value of 2518.47. Momentum, as being one of the technical indicators, ranged between 2862.71 (maximum) and -2634.72 (minimum). Stochastic K% was recorded at 100 as maximum and at 0.9 as minimum, while the Stochastic D% value recorded a maximum of 87.9, minimum of 3.71. Similarly, the Relative Strength Index recorded maximum value of 100, minimum value of 1.01, while the Moving Average Convergence and Divergence earned maximum value of 818.7, minimum value of -1002.7. Williams R% attained zero (0) as the maximum and minimum of -93.9 whereas the A/D Oscillator recorded maximum at 100 and minimum at 1.36. Commodity Channel Index values ranged from a maximum of 311.6 and a minimum of -201.4 during 1999-2007.

Table 5: Data Validation For Bse-Sensex In Pre-Crisis Period (1999-2007)
S. No. Technical Indicators Used Minimum Maximum Mean Std. Dev.
1 Simple Day Moving Average 2600.12 20375.87 6588.31 4120.63
2 Weighted Day Moving Average 2518.47 20142.87 6374.17 3978.15
3 Momentum -2634.72 2862.71 47.14 13.28
4 Stochastic K% 0.9 100 48.21 64.1
5 Stochastic D% 3.71 87.9 42.04 48.6
6 Relative Strength Index 1.01 100 48.16 40.3
7 MACD -1002.7 818.7 41.3 -30.21
8 Larry Williams R% -93.9 0 -32.6 -21.7
9 A/D Oscillator 1.36 100 43.7 83.5
10 Commodity Channel Index -201.4 311.6 21.16 91.7

From Table 6, it is clear that the values of Simple day moving average were at 29681.77 (maximum), 8160.4 (minimum), 19989.02 (mean) and 5089.17 (standard deviation), during the post-crisis period. At the same time, weighted day moving average attained the maximum value of 28173.16 and minimum value of 8074.81. Momentum values ranged between 2462.71 (maximum) and -3834.27 (minimum). Stochastic K% recorded maximum value of 100 and minimum of 1.1 while Stochastic D% was recorded at 97.79 as maximum, 5.17 as minimum. Similarly, the Relative Strength Index of BSE-Sensex registered maximum value of 100 and minimum value of 1.07. The values of Moving Average Convergence and Divergence was ranged between 918.7 (maximum) and -1124.2 (minimum). Williams R% posted zero (0) as maximum and -98.9 as minimum. A/D Oscillator was maximum at 100 and minimum at 1.78. Commodity Channel Index ranged between 381.26 and -243.4 during 2008-2016.

Validation for NSE-Nifty

Table 7: Data Validation For Nse-Nifty In Pre-Crisis Period (1999-2007)
S. No. Technical Indicators Used Minimum Maximum Mean Std. Dev.
1 Simple Day Moving Average 2573.15 6513.25 3618.26 1247.35
2 Weighted Day Moving Average 2138.45 5914.71 3179.69 1072.51
3 Momentum -110.7 687.4 119.91 6.85
4 Stochastic K% 1. 24 91. 6 53.15 62.81
5 Stochastic D% 3.89 89.4 52.7 49.71
6 Relative Strength Index 1.03 100 46.03 40.1 8
7 MACD -276.13 218.3 10.2 -11. 9
8 Larry Williams R% -89.16 -0.68 -36. 4 -20.7
9 A/D Oscillator 1.78 87.24 49.12 80.94
10 Commodity Channel Index -217.3 319. 1 20.4 87. 71

Source: www.nseindia.com (using SPSS 20.0)

Table 7 reveals the statistical values of ten technical parameters, used to validate both training and hold-out dataset, in respect of NSE-Nifty, during 1999-2007. Simple day moving average posted the maximum value of 6513.25, minimum value of 2573.15 and mean value of 3618.26 and standard deviation of 1247.35, during the pre-crisis period. The value of weighted day moving average was registered at 5941.71 as maximum and 2138.45 as minimum. Momentum values ranged between 687.4 (maximum) and -110.7 (minimum). Stochastic K% recorded the maximum value of 91.6 and minimum value of 1.24. Similarly, Stochastic D% attained a value of 89.4 as maximum, 3.89 as minimum. Relative Strength Index posted the maximum value of 100, minimum value of 1.03. Moving average convergence and divergence attained the maximum value of 218.3, minimum value of -276.13. Williams R% earned the maximum value of -0.68 and minimum value of -89.16. But the A/D Oscillator recorded maximum value as 87.24 and the value of minimum as 1.78. Commodity channel index ranged between 319.1 (maximum) and -217.3 (minimum), during the pre-crisis period.

The statistical values of ten technical parameters, used to validate both training and hold-out dataset in respect of NSE-Nifty, for the post-crisis period (2008-2016), are given in Table 8. Simple day moving average had posted the maximum value at 8996.25, minimum value at 5763.21, its mean value at 4186.63 and value of standard deviation at 2417.59. The Weighted day moving average attained the maximum value of 7941.71 and minimum value of 4838.45. Momentum values ranged between 781.4 (maximum) and -128.70 (minimum). Stochastic K% recorded the maximum value of 97.56 and minimum of 1.82. Stochastic D% recorded the maximum value of 97.9 and minimum value of 4.08. Relative Strength Index attained the maximum value of 100, minimum value of 1.4. But the Moving average convergence and divergence recorded the maximum value of 279.1, minimum value of -347.33. Williams R% posted values, ranging from -0.79 to -94.16. A/D Oscillator received the maximum value as 97.24, minimum as 1.89. But, Commodity channel index ranged between 374.21 and -238.5.

Table 8: Data Validation For Nse-Nifty In Post-Crisis Period (2008-2016)
S. No. Technical Indicators Used Minimum Maximum Mean Std. Dev.
1 Simple Day Moving Average 5763.21 8996.52 4186.63 2417. 59
2 Weighted Day Moving Average 4838.45 7914.71 3497.69 2027.51
3 Momentum -128.70 781.4 167.91 7.31
4 Stochastic K% 1.82 97.56 60.51 76.33
5 Stochastic D% 4.08 97.9 60.2 58.71
6 Relative Strength Index 1.4 100 51.40 43.28
7 MACD -347.33 279.1 12.82 -13.19
8 Larry Williams R% -94.16 -0.79 -38. 9 -22.7
9 A/D Oscillator 1.89 97.24 54.31 85.94
10 Commodity Channel Index -238.5 374.21 22.89 94.27

Source: www.nseindia.com (using SPSS 20.0)

From the analysis of Tables 5-8, occurrence of upward and downward movements (bull/bear events) of both stock indices (BSE-Sensex and NSE-Nifty) was relatively similar, in terms of their validation of data, during the period of study (1999-2016). The retail investors are advised to note these trends of stock indices and devise their investment strategies accordingly.

Major Findings

The analysis of Tables 1 and 2 clearly revealed that Observations of Total Dataset (Continuous), both BSE-Sensex and NSE-Nifty, were rampant and robust in terms of both moving forward and backward (upward/downward), from the current closing price points and index movements. The incidence of growth/decline (bull and bear events) had been relatively similar, during both the pre-crisis as well as the post-crisis period. Moreover, the Tables 3 and 4, analysed of parameter setting data (discrete), on both cases (training and hold-out dataset) indicated the fact that both sample indices Sensex and Nifty were found to be highly stochastic. The occurrence of bull and bear events had been relatively similar, during both the pre-crisis and (1999-2007) and the post-crisis period (2008-2016).

Similarly, it is found from Tables 5 and 7, that the values of Momentum, Relative Strength Index, Williams R% and Commodity Channel Index (in terms of Maximum, Minimum, Mean and Standard deviation) were of positive and negative trend synonymously and hence do not accept the null hypothesis NH-1 (There is no corresponding relationship between the upward/downward (growth/decline) movements of BSE-Sensex and NSE-Nifty during the pre-global financial crisis period). The prediction of stock index movements of Sensex and Nifty, has recorded similar index movements (upward/downward) during the pre-crisis period.

The results of Tables 6 and 8, shows that the values of Momentum, Relative Strength Index, Williams R% and Commodity Channel Index (in terms of Maximum, Minimum, Mean and Standard deviation) indicated positive and negative trend to be synonymous for both BSE-Sensex and NSE-Nifty indices. Hence reject NH-2 (There is no corresponding relationship between the upward/downward (growth/decline) movements of BSE-Sensex and NSE-Nifty during post global financial crisis period). The results indicated that the prediction of index movements of Sensex and Nifty, witnessed similar movements (upward and downward) during the post-crisis period. Hence, retail investors should note these trends of stock indices and devise their investment strategies accordingly.

Conclusion

The analysis of this study clearly showed that both BSE-Sensex and NSE-Nifty were rampant and robust, in terms of index movements, moving forward and backward, from the current/existing points of daily closing prices. At the same time, the incidences of stochastic movements of both the indices were found to be similar, during the study period. Further, the values of Momentum, Relative Strength Index, Williams R% and Commodity Channel Index values (in terms of Maximum, Minimum, Mean and Standard Deviation) indicated both positive and negative trend and it implied the prediction of stock index movements of Sensex and Nifty recorded similar trends (upward/downward), during both the pre-crisis and post-crisis periods.

Scope For Future Research

1. Researchers could attempt to forecast sectorial indices of BSE and NSE.

2. Efforts could be made, to study the movements of stock indices of developed economies, DJIA, S&P-500 (USA), Nikkei-225 (Japan) and FTSE-100 (UK).

3. A comparative analysis of global stock indices, with the Indian stock indices, could also be made, by applying the neural network methods.

References

  1. Amitai, E. (1976). Future analysis. Analysen & Prognosen, 19-20.
  2. Andre, C. & Beatriz, V.M.M. (2008). Evaluating the forecast accuracy of emerging market stock returns. Emerging Markets Finance and Trade, 44(1), 21-40.
  3. Defu, Z., Qingshan, J. & Xin, L. (2005). Application of neural networks in financial data mining. World Academy of Science, Engineering and Technology, 1, 136-139.
  4. Fama, E. (1970). Efficient capital markets. Journal of Finance, 25(2), 383-417.
  5. Garg, A., Sriram, S. & Tai, K. (2013). Empirical analysis for model selection criteria for genetic programming in modelling of time series system. IEEE Conference on Computational Intelligence (CIFE).
  6. Hansen, J.V. & Nelson, R.D. (2003). Forecasting and recombining time-series components by using neural networks. Journal of the Operational Research Society, 54(3), 307-317.
  7. Jigar, P., Sahil, S., Priyank, T. & Kotecha, K. (2015a). Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques. Expert Systems with Applications, 42, 259-268.
  8. Jigar, P., Sahil, S., Priyank, T. & Kotecha, K. (2015b). Predicting stock market index using fusion of machine learning techniques. Expert Systems with Applications, 42, 2162-2172.
  9. Kim, K.J. (2003). Financial time series forecasting using support vector machines. Neurocomputing, 55, 307-319.
  10. Melek, A.B. & Derya, A. (2010). An Adaptive Network-Based Fuzzy Inference System (ANFIS) for the prediction of stock market return: The case of the Istanbul stock exchange. Expert Systems with Applications, 37, 7908-7912.
  11. Miao, K., Chen, F. & Zhao, Z.G. (2007). Stock price forecast based on bacterial colony RBF neural network. Journal of Qingdao University (Natural Science Edition), 2, 11.
  12. Nair, B.B., Sai, S.G., Naveen, A., Lakshmi, A., Venktesh, G. & Mohandas, V. (2011). A ga-artificial neural network hybrid system for financial time series forecasting. Information Technology and Mobile Communication, 499-506.
  13. Ou, P. & Wang, H. (2009). Prediction of stock market index movements by ten data mining techniques. Modern Applied Science, 3, 28.
  14. Robert, S. & David, W. (2011). A network value theory of a market and puzzles. Financial Analysts Journal, 67(5), 69-85.
  15. Robin, H. & Ryan, O. (2009). A manipulator can aid prediction market accuracy. Economica, 76(302), 304-314.
  16. Shin, F.W. & Shie, J.L. (2015). Employing local modelling in machine learning based methods for time-series prediction. Expert Systems with Applications, 42, 341-354.
  17. Shuai, W. & Wei, S. (2014). Forecasting direction of China security index 300 movement with least squares support vector machine. Procedia Computer Science, 31, 869-874.
  18. Spanning tree algorithm. Quantitative Economics, 12(2), 44-89.
  19. Teo, J. & Douglas, W. (2006). The profitability of daily stock market indices trades based on neural network predictions: Case study for the S&P 500, the DAX, the TOPIX & the FTSE in1965-1999. Applied Financial Economics, 14(4), 285-297.
  20. Wang, J.H. & Leu, J.Y. (1996). Stock market trend protection using arima-based neural networks. IEEE International Conference of Neural Networks.
Get the App