Efficiency of Indian Commodities Market:

A Study of Agricultural Commodity Derivatives Traded on NCDEX

*Gurbandini Kaur
** Rao, D.N.

1. Introduction

Since commodity “futures” trading was permitted in 2003, the commodity

derivative market in India has witnessed phenomenal growth. Indian commodity
market expanded almost by 50 times in a span of 5 years from Rs 665.30 billion in
2002 to Rs 33,753.36 billion in 2007 registering a Compounded Annual Growth Rate
(CAGR) of a little over 119.3% and is now expected to grow at a steady growth rate
of about 30% by 2010 and touch a volume of Rs 74,156.13 billion due to the
continued active and wide participation of traders (ASSOCHAM findings). [Indian
Rupees One Billion is equivalent to approximately US Dollars 22.63 millions at the
exchange rate of US$1 = INR 44.2]

Trading of commodity derivatives on exchange platforms helps to achieve price

discovery, price risk management besides helping economy with better resource
allocation. Though the volume of commodity futures trade increased exponentially
since its launch in 2003, the functioning of the futures market came under scrutiny
during 2008-2009 due to price rise. The role of futures market in stabilizing spot
prices was widely discussed.

However, the success of these markets in performing the stabilizing function

critically depends on whether they are “efficient” (Fama, 1970, p383).
* Gurbandini Kaur is currently associated with Centre for Management Education,
All India Management Association, New Delhi, India as Assistant Professor.
** Rao, D.N. is currently the President of Kings University, Accra, Ghana and the
former Director of Centre for Management Education, All India Management
Association, New Delhi, India.

Electronic copy available at: http://ssrn.com/abstract=1600687

Moreover, the “dual price system” under which different prices for same
commodities exist, the administration of a part of market for the commodity by the
government is expected to give rise to inefficiency. In an emerging market context
like India, the growth of commodity future market would depend on efficiency of the
future market. An efficient market is one in which the spot market “fully reflects” the
available information and no one can consistently make profits and futures
eliminate the possibility of guaranteed profits [Sahi & Raizada, 2006]. An efficient
market would provide reliable forecasts of spot prices in future. This would also
provide export – import oriented firms with better risk management capabilities.

2. Literature Review

Several studies have empirically tested the reaction of security prices to the release
of different information. Beaver (1968), Foster (1981), Ball and Brown (1968),
Beaver et al (1979) are some of the studies which find significant reaction in the
security prices to the release of corporate events announcement information. On
the Indian stock market, M.Obaidullah (1992), S Srinivasan, and Kakati (2001), Jijo
Lukose and Narayan Rao (2002) are some of the studies which have tested the
efficiency of the Indian stock market with respect to corporate events
announcement information like accounting information, dividend announcement,
bonus announcement, right issue, mergers & acquisition and stock split etc.

There have been a few studies on the performance of Indian commodity derivatives
market. Pravakar (2009) examined the efficiency and future trading price nexus for
five top selected commodities (gold, copper, petroleum crude, soya oil and chana).
Results suggested that the market is efficient for all five commodities. Lokare
(2007) found that although Indian commodity market is yet to achieve minimum
critical liquidity in some commodities (sugar, pepper, gur and groundnut), almost all
the commodities show an evidence of co-integration between spot and future prices
revealing the right direction of achieving the improved operational efficiency,
though at a slower rate. The efficiency of agricultural commodity future market in

Electronic copy available at: http://ssrn.com/abstract=1600687

China was studied by Wang (2005) and Bingfanke (2005) and the results suggested
a long term equilibrium relationship between the futures price and spot price for
Soya beans and weak form of efficiency in short term. The study also highlighted
inefficient future market for wheat and suggested that it may have been caused by
over speculation and government intervention.

There have been a number of studies that have analyzed efficiency of commodity
markets in the developed countries. Andrew, Mathew (2002) studied market
efficiency and bias in agricultural commodities ( live cattle, hogs and soybean). The
efficiency of commodity markets can be analyzed by using approaches of Fama
(1970). Elam and Dixon (1988) have shown the invalidity of conventional F tests
for market efficiency estimation for non stationary time series modeling. Stein
(1991) has estimated the social loss due to inefficiency of the future markets.
Similar study has been conducted for future market in China. The methodology
used in the study has been developed from Lai and Lai (1991).

The econometric techniques developed by Engle and Granger (1987) and Johnsen
and Juselius (1990) for co-integration allow for a Vector Auto Regressive (VAR)
model for determination of spot and future prices and allows for testing market
efficiency.

In the developed as well as developing markets, many studies have been conducted
to test the efficiency of stock markets. Fama (1991) classifies market efficiency into
following forms-weak, semi-strong and strong. Barua (1987), Chan, Gup and Pan
(1997) observed that the major Asian markets were weak form inefficient. Similar
results were found by Dickinson and Muragu (1994) for Nairobi Stock market,
Cheung et al (1993) for Koua and Taiwan and HO and Cheung (1994) for Asian
markets. On the other hand, Groenewold and Kang (1993) found Australian market
semi-strong form efficient.

Electronic copy available at: http://ssrn.com/abstract=1600687

Studies in India primarily focused at examining weak-form efficiency. One of the
studies was by Sharma & Kennedy (1977). They examined Bombay, London & New
York Stock exchanges, testing for weak-form efficiency using run test and spectral
analysis. The results were in support of efficient market hypothesis. Barua (1981)
studied 20 stocks listed in Bombay Stock Exchange (BSE) as well as Financial
Express Index (FE). Yalawar (1988) also examined a number of stocks using
correlation test and run test and supported the hypothesis of efficient market, which
Choudhary (1981) conducting similar tests rejected the hypothesis that the markets
are efficient. Poshakwale (1996) showed that Indian Stock Market was efficient in
weak form; he used daily BSE index data for the period 1987 to 1994.

On the same lines Barman and Madhusoodhan (1991) using unit root test and
variance ratio test, Reddy (1998) and Ahmad et al (2006) using run test and ARCH,
GARCH models, all provides evidence in rejection of the hypothesis of efficient
markets.

Review of literature indicates that much empirical research has been done for
studying efficiency of stock markets in developed and developing countries
including India. However, not much research has been done on testing efficiency of
commodity markets internationally and particularly in the Indian context. Hence,
the paper attempts at testing the efficiency of commodity markets in India with
reference to agricultural derivatives traded on National Commodity Derivatives
Exchange (NCDEX).

3. Objective of the study

To test the market efficiency of select agricultural commodity derivatives in the

Indian context.

4
4. Research Methodology

In India, NCDEX is considered as prime national level commodity exchange for

agricultural commodities and hence selected for the study and the time frame
chosen for the study is the future contracts originating and expiring during the
period July 2008 to July 2009. Four commodities namely - Pepper Malabar, Refined
Soya Oil, Guar seed and Chana have been selected for the study as they account for
almost two-thirds of the volume and value of agricultural commodities traded on
NCDEX. The Future Open and Close, Spot Open and Close prices on all trading days
during the period (January to June of year 2009) were obtained from the home page
of NCDEX (www.ncdex.com).

Hypothesis

(H ): Agricultural commodity market in India is not efficient (in the weak form of
Efficient Market Hypothesis).

(H ): Agricultural commodity market in India is efficient (in the weak form of

Efficient Market Hypothesis).

In all, 27 Future Contracts for the above four commodities were analyzed for the
period of study. Autocorrelation test and Run test are used to test the efficiency of
the agricultural commodity market.

Autocorrelation
Autocorrelation is one of the statistical tools used for measuring the dependence of
successive terms in a given time series. Therefore it is used for measuring the
dependence of successive price changes. It is the basic tool used to test the weak
form of Efficient Market Hypothesis.

Runs Test

A Runs test is another common approach to test for statistical independences but
unlike autocorrelation coefficient, it does not require returns to be normally

5
distributed. The run test is a non-parametric test that is designed to examine
whether successive price changes are independent. The test is based on the premise
that if a series of a data is random, the observed number of runs in the series should
be close to the expected number of runs.

6. Results and Discussions

Autocorrelation analysis

Autocorrelations for spot and future price of select agricultural commodities (for
contracts expiring from Jan to July 2009) with 6 lags for period of 13 months, i.e.
July 2008 to July 2009 (both the months inclusive) has been calculated and the
results are presented in Tables 1 to 4.

The autocorrelation at lag one is lowest for pepper future (open) for contract ending
in the month of April (.407) and highest for spot (close) price for January ending
contract of Pepper (.987).

We find the coefficients are high at lags in the beginning and the values continue to
fall as the lags increase. That is ACF (k) are very high and decline slowly as the lag
value (k) increases. However, the fall in the value of autocorrelation is not much.
This may suggest that information embedded in longer period of lags would be as
influential in determining future price as that of information embedded in short lag
periods. The results show significant evidence of linear dependence of the lags for
the four agricultural commodities studied.

For Refined oil, the autocorrelation coefficient for lag 1 is 0.818, which is very much
larger than twice the standard error (=2*0.121). Thus the autocorrelation differ
significantly from zero. From the table, we can see that out of the 144
autocorrelations computed for the contract ending from January to July for refined
oil, 137 differ significantly from zero. That is 95% of the autocorrelations differ

6
significantly. While for rest three commodities that is for Guar seed, Chana, Pepper
all the values of autocorrelation differ significantly from zero.

The t-values of the autocorrelations corresponding to the future and spot price for
the contract ending from January to July for the select agricultural commodities
were computed. It was found that for three commodities Guar seed , Chana and
Pepper the autocorrelations are significantly different from zero and the
corresponding t-values are greater than 1.96 (at 5% level of significance). Thus the
commodity prices are biased random time series and the commodity market is
efficient in pricing its contracts. However the same result for t value is not found for
refined oil. Seven t values out of the total values computed are found to be less than
1.96.

Further, Run Test was conducted to test the Null hypothesis and the results are
presented in Tables 5 to 8.

Run Test analysis

Further, Run Test was conducted to test the Null hypothesis and the results are
presented in Tables 5 to 8. The total number of runs is a measure of randomness
since too many or too few runs indicate dependence between observations. Number
of observations less than test value denotes the number of observations below
median, while Number of observations equal to or greater than the test value
indicates the number greater than or equal to the median. It is the positive change
when the return is greater than the median, a negative change when the return is
less than the mean and zero when the return equals to median. The test statistic Z is
the observed value.

The results of run test indicate that both future and spot price for all select
agricultural commodities are weak form efficient. The information regarding the
yesterday prices are effectively absorbed by today’s price. The significant negative Z
value for prices indicate that the actual number of runs fall short of the expected

7
number of runs. The negative Z values for prices are also indicative of positive
autocorrelation, which is consistent with the results obtained from Autocorrelation.

Table 1: Results of Autocorrelation for Pepper Derivatives

Pepper
Lags → 1 2 3 4 5 6
Months ↓ Std error
January
F (O) 0.985 0.967 0.947 0.925 0.902 0.88 0.089
t-value 11 10.8 10.6 10.3 10.2 9.8
F (C ) 0.985 0.968 0.948 0.924 0.901 0.877 0.089
t-value 11 10.8 10.6 10.3 10.1 9.8
S (O) 0.986 0.968 0.947 0.926 0.902 0.879 0.089
t-value 11 10.8 10.6 10.4 10.2 9.8
S (C ) 0.987 0.969 0.949 0.927 0.903 0.88 0.089
t-value 11 10.8 10.6 10.4 10.2 9.8
February
F (O) 0.96 0.919 0.88 0.838 0.795 0.756 0.089
t-value 10.7 10.3 9.8 9.4 8.9 8.4
F (C ) 0.964 0.926 0.885 0.84 0.799 0.758 0.089
t-value 10.8 10.4 9.9 9.4 8.9 8.5
S (O) 0.969 0.935 0.869 0.855 0.811 0.768 0.089
t-value 10.8 10.5 9.7 9.6 9.1 8.6
S (C ) 0.97 0.934 0.897 0.854 0.81 0.769 0.089
t-value 10.8 10.4 10 9.5 9.1 8.6
March
F (O) 0.956 0.919 0.883 0.848 0.806 0.763 0.089
t-value 10.7 10.3 9.9 9.5 9 8.5
F (C ) 0.963 0.927 0.893 0.855 0.816 0.772 0.089
t-value 10.8 10.4 10 9.6 9.1 8.6
S (O) 0.962 0.925 0.887 0.85 0.807 0.762 0.089
t-value 10.8 10.3 9.9 9.5 9 8.5
S (C ) 0.968 0.93 0.897 0.857 0.812 0.767 0.089
t-value 10.8 10.4 9.7 9.6 9.1 8.6
April
F (O) 0.407 0.365 0.375 0.323 0.282 0.24 0.089
t-value 4.5 4.1 4.2 3.6 3.1 2.6
F (C ) 0.913 0.817 0.732 0.636 0.551 0.455 0.089
t-value 10.2 9.1 8.2 7 6.1 5.1
S (O) 0.921 0.842 0.765 0.68 0.591 0.497 0.089
t-value 10.3 9.4 8.5 7.6 6.6 5.5
S (C ) 0.925 0.839 0.762 0.673 0.585 0.491 0.089
t-value 10.3 10.5 8.5 7.5 6.5 5.5
May
F (O) 0.949 0.892 0.842 0.787 0.728 0.664 0.089

8
 Pepper
Lags → 1 2 3 4 5 6
Months ↓ Std error
t-value 10.6 10 9.4 8.8 8.8 8.1
F (C ) 0.957 0.901 0.847 0.792 0.736 0.679 0.089
t-value 10.7 10.1 9.5 8.8 8.2 7.6
S (O) 0.972 0.934 0.894 0.844 0.788 0.731 0.089
t-value 10.9 10.4 10 9.4 8.8 8.2
S (C ) 0.972 0.933 0.895 0.846 0.792 0.735 0.089
t-value 10.9 10.4 10 9.5 8.8 8.2
June
F (O) 0.436 0.426 0.391 0.385 0.338 0.325 0.089
t-value 4.8 4.7 4.3 4.3 3.7 3.6
F (C ) 0.951 0.894 0.841 0.78 0.722 0.667 0.089
t-value 10.6 10 9.4 8.7 8.1 7.4
S (O) 0.969 0.923 0.873 0.815 0.759 0.703 0.089
t-value 10.8 10.3 9.8 9.1 8.5 7.8
S (C ) 0.97 0.923 0.873 0.815 0.76 0.703 0.089
t-value 10.8 10.3 9.8 9.1 8.5 7.8
July
F (O) 0.949 0.917 0.868 0.817 0.759 0.704 0.089
t-value 10.6 10.3 9.7 9.1 8.5 7.9
F (C ) 0.959 0.914 0.869 0.811 0.752 0.695 0.089
t-value 10.7 10.2 9.7 9.1 8.4 7.8
S (O) 0.976 0.946 0.911 0.868 0.826 0.778 0.089
t-value 10.9 10.6 10.2 9.7 9.2 8.7
S (C ) 0.978 0.946 0.91 0.868 0.824 0.776 0.089
t-value 10.9 10.6 10.2 9.7 9.2 8.7

S (O): Spot Open; S (C ): Spot Close; F (O): Futures Open; F (C): Futures Close

Table 2: Results of Autocorrelation for Refined Soya Oil Derivatives

Refined Soya Oil

Lags → 1 2 3 4 5 6
Months ↓ Std error
January
F (O) 0.818 0.68 0.556 0.509 0.461 0.437 0.121
t-value 6.7 5.6 4.5 4.2 3.8 3.6
F (C ) 0.844 0.655 0.65 0.545 0.522 0.459 0.121
t-value 6.9 5.4 5.4 4.5 4.3 3.7
S (O) 0.713 0.304 0.056 0.017 0.118 0.251 0.121
t-value 5.8 2.5 0.46 0.14 0.97 2.07
S (C ) 0.685 0.228 0.014 0.016 0.136 0.264 0.121

9
 Refined Soya Oil
Lags → 1 2 3 4 5 6
Months ↓ Std error
t-value 5.6 1.8 0.11 0.13 1.12 2.18
February
F (O) 0.843 0.682 0.577 0.507 0.458 0.389 0.121
t-value 6.9 5.6 4.7 4.1 3.7 3.2
F (C ) 0.868 0.697 0.585 0.535 0.491 0.422 0.121
t-value 7.1 5.7 4.8 2.5 4.5 3.4
S (O) 0.852 0.641 0.471 0.397 0.392 0.41 0.121
t-value 7.04 5.2 3.8 3.2 3.2 3.3
S (C ) 0.853 0.638 0.482 0.404 0.397 0.409 0.121
t-value 7.04 5.2 3.8 3.3 3.2 3.3
March
F (O) 0.798 0.647 0.528 0.446 0.421 0.338 0.121
t-value 6.5 5.3 4.3 3.6 3.4 2.7
F (C ) 0.849 0.663 0.521 0.445 0.397 0.33 0.121
t-value 7.01 5.4 4.3 3.6 3.2 2.7
S (O) 0.911 0.808 0.727 0.693 0.681 0.665 0.121
t-value 7.5 6.6 6 5.7 5.6 5.4
S (C ) 0.918 0.81 0.739 0.705 0.69 0.667 0.121
t-value 7.5 6.6 6.1 5.8 5.7 5.5
April
F (O) 0.837 0.695 0.56 0.442 0.32 0.245 0.121
t-value 6.9 5.7 4.6 3.6 2.6 2.02
F (C ) 0.851 0.7 0.566 0.444 0.348 0.263 0.121
t-value 7.03 5.7 4.6 3.6 2.8 2.17
S (O) 0.882 0.767 0.654 0.555 0.472 0.391 0.121
t-value 7.2 6.3 5.4 4.5 3.9 3.2
S (C ) 0.893 0.78 0.672 0.575 0.492 0.408 0.121
t-value 7.3 6.4 5.5 4.7 4.06 3.3
May
F (O) 0.937 0.901 0.857 0.823 0.777 0.725 0.121
t-value 7.7 7.4 7.08 6.8 6.4 5.9
F (C ) 0.966 0.927 0.881 0.843 0.802 0.759 0.121
t-value 7.9 7.6 7.2 6.9 6.6 6.2
S (O) 0.96 0.909 0.86 0.821 0.783 0.734 0.121
t-value 7.9 7.5 7.1 6.7 6.4 6.06
S (C ) 0.963 0.917 0.87 0.836 0.796 0.749 0.121
t-value 7.9 7.5 7.1 6.9 6.5 6.1

June
F (O) 0.944 0.886 0.826 0.775 0.734 0.685 0.121
t-value 7.8 7.3 6.8 6.4 6.06 5.6
F (C ) 0.941 0.873 0.809 0.756 0.719 0.672 0.121
t-value 7.7 7.2 6.6 6.2 5.9 5.5
S (O) 0.95 0.887 0.82 0.767 0.713 0.645 0.121

10
 Refined Soya Oil
Lags → 1 2 3 4 5 6
Months ↓ Std error
t-value 7.8 7.3 6.7 6.3 5.8 5.3
S (C ) 0.952 0.892 0.827 0.78 0.726 0.664 0.121
t-value 7.8 7.3 6.8 6.4 6 5.2
July
F (O) 0.925 0.851 0.781 0.725 0.656 0.594 0.121
t-value 7.6 7.03 6.4 5.9 5.4 4.9
F (C ) 0.925 0.853 0.767 0.7 0.653 0.599 0.121
t-value 7.6 7.04 6.3 5.7 5.3 4.9
S (O) 0.952 0.893 0.843 0.8 0.754 0.708 0.121
t-value 7.8 7.3 6.9 6.6 6.2 5.8
S (C ) 0.953 0.9 0.845 0.806 0.762 0.715 0.121
t-value 7.8 7.4 6.9 6.4 6.2 5.9

Table 3: Results of Autocorrelation for Guar seed

Guar seed
Lags → 1 2 3 4 5 6
Months Std error
↓
January
F (O) 0.972 0.942 0.91 0.879 0.841 0.804 0.086
t-value 11.3 10.9 10.58 10.2 9.7 9.3
F (C ) 0.971 0.942 0.911 0.873 0.836 0.797 0.086
t-value 11.2 10.9 10.5 10.1 9.7 9.2
S (O) 0.966 0.93 0.893 0.858 0.812 0.775 0.086
t-value 11.2 10.81 10.3 9.9 9.4 9.01
S (C ) 0.97 0.931 0.895 0.855 0.812 0.771 0.086
t-value 11.2 10.8 10.4 9.9 9.4 8.9
March
F (O) 0.96 0.92 0.881 0.835 0.789 0.751 0.086
t-value 11.1 10.6 10.2 9.7 9.1 7-Aug
F (C ) 0.961 0.922 0.878 0.828 0.78 0.739 0.086
t-value 11.1 10.6 10.2 9.6 9.06 8.5
S (O) 0.924 0.842 0.759 0.673 0.575 0.516 0.086
t-value 10.7 9.7 8.8 7.8 6.6 6
S (C ) 0.934 0.85 0.763 0.674 0.585 0.517 0.086
t-value 10.8 9.8 8.8 7.8 6.8 6.01
April
F (O) 0.896 0.786 0.676 0.552 0.45 0.382 0.086
t-value 10.4 9.1 7.8 6.4 5.2 4.4
F (C ) 0.886 0.788 0.694 0.575 0.486 0.408 0.086
t-value 10.3 9.1 8.06 6.6 5.6 4.7
S (O) 0.916 0.832 0.736 0.637 0.528 0.446 0.086

11
 Guar seed
Lags → 1 2 3 4 5 6
Months Std error
↓
t-value 10.6 9.6 8.5 7.4 6.1 5.1
S (C ) 0.918 0.83 0.75 0.654 0.544 0.459 0.086
t-value 10.6 9.6 8.7 7.6 6.3 5.3
May
F (O) 0.95 0.888 0.826 0.766 0.708 0.661 0.086
t-value 11.04 10.3 9.6 8.9 8.2 7.6
F (C ) 0.934 0.881 0.827 0.764 0.706 0.662 0.086
t-value 10.8 10.2 9.6 8.8 8.2 7.6
S (O) 0.957 0.913 0.866 0.821 0.775 0.734 0.086
t-value 11.1 10.6 10.06 9.5 9.01 8.5
S (C ) 0.957 0.913 0.869 0.824 0.778 0.739 0.086
t-value 11.1 10.6 10.1 9.5 9.04 8.5
June
F (O) 0.936 0.869 0.802 0.744 0.677 0.624 0.086
t-value 10.8 10.1 9.3 8.6 7.08 7.2
F (C ) 0.933 0.872 0.811 0.742 0.687 0.633 0.086
t-value 10.8 10.1 9.4 8.6 7.9 7.3
S (O) 0.964 0.927 0.886 0.852 0.816 0.785 0.086
t-value 11.2 10.7 10.3 9.9 9.4 9.1
S (C ) 0.963 0.927 0.889 0.854 0.818 0.787 0.086
t-value 11.1 10.7 10.3 9.9 9.5 9.1
July
F (O) 0.905 0.825 0.746 0.689 0.633 0.584 0.086
t-value 10.5 9.5 8.6 8.01 7.3 6.7
F (C ) 0.909 0.842 0.774 0.711 0.658 0.606 0.086
t-value 10.5 9.7 8.9 8.2 7.6 7.6
S (O) 0.955 0.913 0.867 0.823 0.79 0.754 0.086
t-value 11.1 10.6 10.08 9.5 9.1 8.7
S (C ) 0.95 0.913 0.869 0.824 0.79 0.755 0.086
t-value 11.04 10.6 10.1 9.7 9.1 8.7

Table 4: Results of Autocorrelation for Chana

Chana
Lags → 1 2 3 4 5 6
Months ↓ Std error
January
F (O) 0.656 0.53 0.404 0.379 0.302 0.231 0.089
t-value 7.3 5.9 4.5 4.5 4.2 2.5
F (C ) 0.746 0.555 0.491 0.377 0.247 0.206 0.089
t-value 8.3 6.2 5.5 4.2 2.7 2.3

12
 Chana
Lags → 1 2 3 4 5 6
Months ↓ Std error
S (O) 0.891 0.76 0.65 0.589 0.531 0.47 0.089
t-value 10 8.5 7.3 6.6 5.9 5.2
S (C ) 0.892 0.763 0.67 0.608 0.528 0.461 0.089
t-value 10 8.5 7.5 6.8 5.9 5.1
February
F (O) 0.95 .9.7 0.858 0.831 0.798 0.761 0.089
t-value 10.6 10.8 9.6 9.3 8.9 8-Jan
F (C ) 0.964 0.921 0.882 0.841 0.807 0.776 0.089
t-value 10.8 10.3 9.9 9.4 9 8.7
S (O) 0.885 0.742 0.615 0.575 0.545 0.506 0.089
t-value 10.9 8.3 6.9 6.4 6.1 5.6
S (C ) 0.897 0.762 0.66 0.612 0.564 0.517 0.089
t-value 10 8.5 7.4 6.8 6.3 5.8
March
F (O) 0.967 0.932 0.893 0.862 0.833 0.804 0.089
t-value 10.8 10.4 10 9.6 9.3 9
F (C ) 0.971 0.937 0.902 0.873 0.841 0.809 0.089
t-value 10.9 10.5 10.2 9.8 9.4 9
S (O) 0.872 0.727 0.599 0.599 0.522 0.464 0.089
t-value 9.7 8.1 6.7 6.7 5.8 5.2
S (C ) 0.886 0.744 0.64 0.592 0.54 0.476 0.089
t-value 9.9 8.3 7.1 6.6 6 5.3
April
F (O) 0.96 0.917 0.875 0.843 0.809 0.771 0.089
t-value 10.7 10.3 9.8 9.4 9 8.6
F (C ) 0.966 0.927 0.885 0.849 0.815 0.778 0.089
t-value 10.8 10.4 9.9 9.5 9.1 8.7
S (O) 0.867 0.716 0.575 0.522 0.476 0.424 0.089
t-value 9.7 8 6.4 5.8 5.3 4.7
S (C ) 0.883 0.736 0.617 0.556 0.5 0.441 0.089
t-value 9.9 8.2 6.9 6.2 5.6 4.9
May
F (O) 0.961 0.921 0.881 0.838 0.79 0.744 0.089
t-value 10.7 10.3 9.8 9.4 8.8 8.3
F (C ) 0.966 0.927 0.888 0.844 0.795 0.748 0.089
t-value 10.8 10.4 9.9 9.4 8.9 8.4
S (O) 0.888 0.766 0.654 0.614 0.581 0.542 0.089
t-value 9.9 8.6 7.3 6.8 6.5 6
S (C ) 0.886 0.744 0.64 0.592 0.54 0.476 0.089
t-value 9.9 8.3 7.1 6.6 6 5.3
June
F (O) 0.948 0.9 0.848 0.797 0.742 0.693 0.089
t-value 10.6 10.1 9.5 8.9 8.3 7.7
F (C ) 0.953 0.902 0.853 0.805 0.748 0.696 0.089

13
 Chana
Lags → 1 2 3 4 5 6
Months ↓ Std error
t-value 10.7 10.1 9.5 9 8.4 7.8
S (O) 0.872 0.754 0.638 0.593 0.541 0.497 0.089
t-value 9.7 8.4 7.1 6.6 6 5.5
S (C ) 0.888 0.773 0.678 0.623 0.57 0.521 0.089
t-value 9.9 8.6 7.6 7 6.4 5.8
July
F (O) 0.955 0.92 0.887 0.854 0.818 0.787 0.089
t-value 10.7 10.3 9.9 9.5 9.1 8.6
F (C ) 0.965 0.93 0.893 0.858 0.822 0.779 0.089
t-value 10.8 10.4 10 9.6 9.2 8.7
S (O) 0.899 0.828 0.724 0.649 0.542 0.462 0.089
t-value 10.1 9.3 8.1 7.2 6 5.1
S (C ) 0.891 0.831 0.733 0.665 0.563 0.487 0.089
t-value 10 9.3 8.2 7.3 6.3 4.9

14
 Table 5: Results of Run Test for Pepper

Pepper
Period Total No. of Test value No. of No. of No. of Z-statistic
observations observations observations Runs
Less than equal to or
Test Value greater than
Test Value
January
F (O) 126 1311.5 63 63 10 -9.6
F (C ) 126 13055.5 63 63 8 -10
S (O) 126 13155.6 63 63 4 -10.7
S (C ) 126 13242.5 63 63 4 -10.7
February
F (O) 122 11901.5 61 61 10 -9.4
F (C ) 122 11935.5 61 61 12 -9.09
S (O) 122 11951.7 61 61 10 -9.4
S (C ) 122 11937 61 61 8 -9.8
March
F (O) 124 11551 62 62 14 -8.8
F (C ) 124 11553.5 62 62 12 -9.1
S (O) 124 11622.3 62 62 8 -9.9
S (C ) 124 11650.7 62 62 8 -9.9
April
F (O) 120 11507 60 60 13 -8.8
F (C ) 120 11489 60 60 9 -9.5
S (O) 120 11547.4 60 60 11 -9.1
S (C ) 120 11493.6 60 60 13 -8.8
May
F (O) 116 11651.5 58 58 11 -8.9
F (C ) 116 11678 58 58 9 -9.3
S (O) 116 11611.6 58 58 5 -10
S (C ) 116 11640.6 58 58 5 -10
June
F (O) 116 12210.5 58 58 17 -7.8
F (C ) 116 12276.5 58 58 12 -8.7
S (O) 116 12162.1 58 58 8 -9.5
S (C ) 116 12152.4 58 58 6 -9.8
July
F (O) 105 12541 52 53 15 -7.5
F (C ) 105 12561 52 53 15 -7.5
S (O) 105 12531.2 52 53 7 -9.1
S (C ) 105 12530.5 52 53 8 -8.9

15
 Table 6: Results of Run Test for Refined Soya Oil

Refined Soya Oil

Period Total No. of Test value No. of No. of No. of Z-statistic
observations observations observations Runs
Less than equal to or
Test Value greater than
Test Value
January
F (O) 32 472.7 16 16 6 -3.7
F (C ) 32 471.2 16 16 8 -3
S (O) 32 491.2 16 16 6 -3.7
S (C ) 32 491.2 16 16 6 -3.7
February
F (O) 54 460 26 28 11 -4.6
F (C ) 54 460.3 27 27 9 -5.2
S (O) 54 486.1 27 27 9 -5.2
S (C ) 54 485.6 27 27 9 -5.2
March
F (O) 68 11551 33 35 18 -4.1
F (C ) 68 11553.5 34 34 16 -4.6
S (O) 68 11622.3 34 34 4 -7.5
S (C ) 68 11650.7 34 34 4 -7.5
April
F (O) 64 11507 32 32 9 -6
F (C ) 64 11489 32 32 11 -5.5
S (O) 64 11547.4 32 32 7 -6.5
S (C ) 64 11493.6 32 32 7 -6.5
May
F (O) 64 454.7 32 32 5 -8.9
F (C ) 64 455.7 32 32 5 -9.3
S (O) 64 468.5 32 32 3 -10
S (C ) 64 466.9 32 32 3 -10
June
F (O) 67 485.1 33 34 7 -6.7
F (C ) 67 484.1 33 34 9 -6.2
S (O) 67 482.7 33 34 7 -6.7
S (C ) 67 482 33 34 7 -6.7
July
F (O) 68 490.8 34 34 10 -6.1
F (C ) 68 489.6 34 34 11 -5.8
S (O) 68 482.7 34 34 6 -7.0
S (C ) 68 481.8 34 34 6 -7

16
 Table 7: Results of Run Test for Guar seed

Guar seed
Period Total No. of Test value No. of No. of No. of Z-statistic
observations observations observations Runs
Less than equal to or
Test Value greater than
Test Value
January
F (O) 127 1742 62 65 6 -10.4
F (C ) 127 1741 63 64 6 -10.4
S (O) 127 1667.4 63 64 10 -9.7
S (C ) 127 1658.1 60 67 12 -9.3
March
F (O) 125 1635 62 63 8 -9.9
F (C ) 125 1625 62 63 8 -9.9
S (O) 125 1597.8 62 63 17 -8.3
S (C ) 124 1601.1 62 62 17 -8.2
April
F (O) 104 1625.5 52 52 15 -7.4
F (C ) 104 1626 52 52 15 -7.4
S (O) 104 1595.1 52 52 17 -7.0
S (C ) 104 1598.5 52 52 15 -7.4
May
F (O) 93 1660 45 48 10 -7.8
F (C ) 93 1658 45 48 10 -7.8
S (O) 93 1606.6 46 47 16 -6.5
S (C ) 93 1606 46 47 14 -6.9
June
F (O) 113 1706 56 57 10 -8.9
F (C ) 113 1714 56 57 14 -8.2
S (O) 113 1619.9 56 57 4 -10.1
S (C ) 113 1619.9 56 57 8 -9.3
July
F (O) 132 1761.5 66 66 12 -9.6
F (C ) 132 1755.5 66 66 14 -9.2
S (O) 132 1742.6 66 66 6 -10.6
S (C ) 132 1744.1 66 66 6 -10.6

17
 Table 8: Results of Run Test for Chana

Chana
Period Total No. of Test value No. of No. of No. of Z-statistic
observations observations observations Runs
Less than equal to or
Test Value greater than
Test Value
January
F (O) 39 2105 19 20 9 -3.5
F (C ) 39 2102 17 22 11 -2.8
S (O) 39 2133.1 19 20 5 -4.8
S (C ) 39 2139 19 20 5 -4.8
February
F (O) 65 2130 32 33 7 -6.6
F (C ) 65 2121 32 33 7 -6.6
S (O) 65 2200 30 35 11 -5.6
S (C ) 65 2200 30 35 9 -6.1
March
F (O) 86 2200 43 43 7 -8
F (C ) 86 2210 42 44 5 -8.4
S (O) 86 2170 43 43 12 -6.9
S (C ) 86 2175 41 45 10 -7.3
April
F (O) 109 2240 54 55 10 -8.7
F (C ) 109 2238 54 55 12 -8.3
S (O) 109 2169.4 54 55 15 -7.7
S (C ) 109 2175 53 56 11 -8.5
May
F (O) 125 2285 60 65 15 -8.7
F (C ) 125 2284 61 64 13 -9.1
S (O) 125 2189.4 62 63 15 -8.7
S (C ) 235 2187 62 63 13 -9
June
F (O) 127 2329 63 64 15 -8.8
F (C ) 127 2323 63 64 13 -9.1
S (O) 127 2169.4 63 64 15 -8.8
S (C ) 127 2175 63 64 13 -9.1
July
F (O) 126 2335.5 63 63 13 -9.1
F (C ) 126 2354.5 63 63 15 -8.7
S (O) 126 2175 59 67 17 -8.3
S (C ) 126 2175 58 68 13 -9.1

18
7. Conclusion

Theoretical basis of the weak- form efficient hypothesis states that the successive
price are independent and past prices have no predictive content to forecast
commodity price. The empirical studies on Efficient Market Hypothesis in emerging
markets show mixed evidence. Based on the theoretical and empirical literature
that is reviewed in this study, the weak form of Efficient Market Hypothesis in the
context of an emerging commodity market namely NCDEX has been investigated.

The study examined the random walk hypothesis and tests the weak form efficiency
of the four major agricultural commodity futures traded on NCDEX using daily data
of trading price for the period of 13 months. We have examined the weak form
efficient market hypothesis over 6 lag period in order to analyze whether the
NCDEX exhibits a trend of market efficiency overtime. Two different statistical tools
namely serial correlation test and run test were used in this study to test commodity
market to investigate the random walk hypothesis.

The empirical results of the study indicate significant evidence of linear dependence
for all four agricultural commodities. This is reflected by high coefficient values of
autocorrelation over the lags. Though the coefficient value is very high at lags in the
beginning and the value goes on falling as we increase the number of lags. The t –
values of the autocorrelation corresponding to future and spot price shows that for
three commodities Guar seed, Chana and Pepper the autocorrelation are
significantly different from zero and corresponding t value is greater than 1.96.
However the same was not observed for refined oil. The results lead to rejection of
Null Hypothesis and acceptance of Alternate Hypothesis that the Commodity
markets are weak form efficient.
The Non parametric run test for the full sample period indicated that both future
and spot price for all select agricultural commodities are efficient in weak form.

19
 References

Ahmad K M, S Ashraf and S.Ahmad (2006), “Testing weak form efficiency for Indian stock
markets,” Economic & Political weekly, pp.49-56.

Andrew M Mckenzine, Mathew T Holt (2002), Market Efficiency in Agricultural Future Markets,
Applied Economics, Vol.34, Issue 12, pp.1519-1532.

Barman, R and T.Madhusoodhan (1991), “Inefficiency and specialization in the Indian securities
market” Vikalpa Vol.16, No.4, pp.17-21.

Barua S (1981), Short run price behaviour securities: some evidence of Indian capital market,
Vikalpa Vol.6, No.2, pp.93-100.

Barua, S K (1987), “Some observations on the report of the High Powered Committee on the
stock exchange reforms”, Annual Issue of ICFAI, Dec.

Beaver, W.H.(1968), “The information content of Annual earnings announcements”, Journal of

Accounting Research, Supplement, Vol.6, pp.67-97.

Chan, K.C. Gup, B.E. and Pan, M S (1997), “International Stock Market Efficiency and Integration:
A study of Eighteen Nations, “Journal of Business Finance & Accounting, Vol.24, No. 6

Cheung, K L Wang, K A and H O Y K (1993), “The pricing of Risky assets Korea and Taiwan,
Applied Financial Economics, Issues 4, Dec,.pp.315-324.

Choudhary, S K (1991), Short-run price behaviour: New evidence on weak form of market
efficiency. Vikalpa, Vol16, No.4, pp.17-21.

Dickinson, J.P., Muragu K (1994) “Market Efficiency in Developing Countries: A Case study of the
Nairobi Stock Exchange, Journal of Business Finance & Accounting, Vol.21, No.1, pp.133-150.

Elam, E and B L Dixon (1988), “Examining the validity as a Test of Future Market Efficiency”,
The Journal of Future Markets, Vol.8, pp.363-372.

Engle, R F and Granger, C (1987), “Co-integration and error correlation representation,

estimation and testing” Econometric Vol.55, pp.251-76.

Fama, E (1991), Efficient Capital Market II, Journal of Finance, Vol.46, pp.1575-1671.

Fama, E.F. (970), “Efficient Capital Market: A Review of Theory and Empirical Work”, The
Journal of Finance, Vol.25, pp.383-417.

Groenewold, N and Kang K C (1993), “The Semi-strong efficiency of the Australian Share Market
Economic record, Vol.69, Issue 207, Dec. pp.405-410.

H O, R Y and Cheung Y.L (1994) “Seasonal Pattern in Volatility in Asia Stock Markets”, Applied
Financial Economics, Vol.4, pp.61-67.

20
Kakati M (2001), “Price Performance of Bonus issue,” Finance Index, Vol. XV, Nov-Dec, pp.1183-
1190.

Kalawar, Y.B. (1988), Bombay Stock Exchange: Rates of return and efficiency. Indian Economic
Journal Vol.35, No.4, pp.68-121.

Lai K.S. and M.Lai (1991), “A co-integration Test for Market Efficiency”, The Journal of Future
Markets, pp.567-575.

Lokare, S.M., 2007, Commodity Derivatives and Price Risk Management: An Empirical Anecdote
from India, Reserve Bank of India, Occasional Papers, vol.28, No.2

Narayanan Rao S (2002),”Market reaction to stock splits – An Empirical Study”, The ICFAI
Journal of Applied Finance, Vol 8, No.2, pp.26-40.

Obaidullah M (1992),”Stock prices adjustment to half-yearly earnings announcement – A test of

market efficiency”, Chartered Accountant, Vol.38, pp.922-924.

Poshakwale, S.(1996), “Evidence on weak form efficiency and Day of the week effect in Indian
stock market, Finance India, Vol.10, No.3, pp.605-616.

Pravakar Sahoo (2009), Efficiency and future trading – Price Nexus in Indian Commodity Future
Market, Global Business Review Vol.10, No.2, pp.187-201.

Reddy Y.S. (1998), “Indian Capital Markets: Theories and Empirical Evidence, Chapter Efficiency
of Indian Stock Market: An empirical analysis of weak-form EMH of the BSE, pp.91-115.

Sahi, G.S. and G Raizada(2006), “Commodity Future Market Efficiency in India and Effect in
Inflation” Social Science Research Network, http://ssrn.com/abstract=949161.

Sharma J L and Robert E. Kennedy (1977), “A Comparative Analysis of Stock Price Behaviour on
the Bombay, London & New York Stock Exchange”, Journal of Financial & Quantitative Analysis
Sep.1977.

Srinivasan (1997),”Security prices Behaviour Associated with Right issues – related events”,
The ICFAI Journal of Applied Finance, Vol.3, No.3, pp.71-81.

Stein, J L (1991), “An evaluation of the performance in speculative markets”, In commodity

Future and Financial Markets”, pp.153-78.

Wang Hong & Bingfanke (2005), Efficiency Test of Agricultural commodity future markets in
china, Australian Journal of Agricultural and Resource Economics, vol.49, No.2, pp.125-141,
June 2005.

21