Predicting Wheat Futures Prices in India

Derivatives are an extremely efficient tool for risk management

“Derivatives are financial weapons of mass destruction”

–Warren Buffett


Futures markets perform their economic roles of price discovery and hedging only when they are efficient. One of the important features of efficient market is that one cannot make abnormal profits from the futures markets by trading in it. This paper addresses the question of whether Indian wheat futures prices can be forecast. This would add to our knowledge whether wheat futures market is efficient, and would enable brokers, traders and speculators to develop profitable trading strategy. We employ the economic variable model to predict the wheat futures prices, and employ out of sample point forecasts. We also evaluate the robustness of our results by employing several alternative specifications, viz. ARMA process and artificial neural network technique. We then test the statistical significance of point forecast using the Diebold and Mariano test. We consider random walk orecast as the bench mark. In order to predict the evolution of wheat futures prices, we use traders’ expectations about the futures prices, a number of economic variables and futures prices (lagged) of wheat. The study finds that the futures price of wheat cannot be forecast, and the wheat futures market is efficient.

This is a preview of subscription content, access via your institution.


  1. 1.

    In Indian context, wheat is the staple food grain crop, and a significant commodity beside rice. The volume of wheat futures contract has gone through a roller coaster ride since its inception on June 10, 2005 at the National Commodity and Derivative Exchange (NCDEX), Mumbai. The government of India banned futures trading in wheat in February 2007 and again lifted the ban in May 2009 (Government of India 2011). The present study has taken the data for the period between May 2009 to August 2014. The average daily volume of wheat futures traded on national exchange decreased by more than 50 percent over the sample period. In 2009, more than 250 futures contracts were traded on average each day, while below 100 contracts were traded averagely on a daily basis in 2014 (calculation based on data from the NCDEX website). This might be because government by interfering (through minimum support price, public distribution system, etc.) in the market, controls its demand and supply so that prices do not inflate.

  2. 2.

    Statistical technique means that we make a forecast based on the past prices only. However, economic metric technique implies that we require the data for the various type of traders who actually trade in the market. We look at the prices they receive, costs they incur, opportunity cost of investing, then we calculate the overall profit of these investors in the market. If they make abnormal profits, we conclude that markets can be predicted.

  3. 3.

    Financialization of commodity markets implies that the flows of the financial investors have impact on the agricultural commodity futures prices.

  4. 4.

    Real rate of interest is the nominal interest (call money) rate minus inflation. Monthly data for whole sale price index is available, so we have calculated inflation for the month and used it for all days in that particular month.

  5. 5.

    This manuscript is part of my Ph.D. dissertation. The work was started in 2015, so data employed in this study is till August 2014. Over the years, we have developed the model and incorporated latest econometric techniques for the study. Though the data is relatively old, but the relevance of the study is quite significant. In addition, we have taken the data from May 21, 2009 because futures trading in wheat started in 2005, and Government of India stopped trading in wheat from August 20, 2007 to May 20, 2009 (Government of India 2008).

  6. 6.

    In case of wheat futures, warehouse stocks plays a dominant role. Despite that we have not selected the variable for predicting grain futures. We have taken the daily data for futures prices in our study. Warehouse stocks data is announced for a year, then it remains constant for a given year. It won’t have any impact as warehouse stocks variable is not changing on daily basis.

  7. 7.

    The results of two tests, i.e., ADF and DF-GLS tests to validate the presence of unit root in the data series have been reported.

  8. 8.

    Wheat and gram have substitutability in production relationship because these two crops are grown in the same season in a given state.

  9. 9.

    Based on the models that we have selected.

  10. 10.

    The terms neurons and nodes are used interchangeably here.

  11. 11.

    The summation formula at a neuron is

    $$I_{j} = \mathop \sum \limits_{j = f}^{s} W_{ij} O_{i}$$

    where \(I_{j}\) is the inputs (weighted) received by neuron \(j\), \(W_{ij}\) is the weight from neuron \(i\). to neuron \(j\), \(O_{i}\) is the signal given by neuron \(i\), \(j\) is the neuron that receives signal, \(i\) the neuron that sends signal and \(f, s\) are the first and last neurons that send signals, respectively.

  12. 12.

    The transfer function is nonlinear. After we calculate \(I_{j}\), we send the strength of the incoming signals to the receiving neuron. They are then transformed and become the outgoing signal to the next layer of neurons. We employ sigmoid function to transform the signal.

    $$O_{j} = \frac{1}{{\left( {1 + e^{{ - I_{j} }} } \right)}}$$

    Since the output of the sigmoid transfer function lies between 0 and 1, \(O_{j}\) lies between 0 and 1. High value of \(I_{j}\) implies that \(O_{j}\) will approach to 1. If \(I_{j}\) is very very negative, the \(O_{j}\) will tend to 0.

  13. 13.

    There is convergence of the training data set, and we get the global minimum point of the RMS value.


  1. Ali, J., & Gupta, B. K. (2011). Efficiency in agricultural commodity futures markets in India: Evidence from cointegration and causality tests. Agricultural Finance Review, 71(2), 162–178.

    Article  Google Scholar 

  2. Bessembinder, H., & Chan, K. (1992). Time-varying risk premia and forecastable returns in futures markets. Journal of Financial Economics, 32(2), 169–193.

    Article  Google Scholar 

  3. Carter, C. A. (2003). Futures and options markets: An introduction. New Jersey: Prentice Hall.

    Google Scholar 

  4. Choudhry, T. (2009). Short-run deviations and time-varying hedge ratios: Evidence from agricultural futures markets. International Review ofFinancial Analysis, 18(1–2), 58–65.

    Article  Google Scholar 

  5. Crain, S. J., & Lee, J. H. (1996). Volatility in wheat spot and futures markets 1950–1993: Government farm programs, seasonality and causality. Journal of Finance, 51(1), 325–343.

    Article  Google Scholar 

  6. DeLurgio, S. A. (1998). Forecasting Principles and Applications. Irwin Professional Publishing.

  7. Diebold, F. X., & Mariano, R. S. (1995). Comparing predictive accuracy. Journal of Business and Economic Statistics, 13(3), 253–63.

    Google Scholar 

  8. Dua, P., Raje, N. & Sahoo, S. (2003). Interest rate modelling and forecasting in India, development research group, No. 24, Reserve bank of India, Mumbai.

  9. Edwards R. Franklin, & Ma, W. C. (1992). Futures and Options. USA: McGraw-Hill Inc.

  10. Elliott, G., Rothenberg, T. J., & Stock, J. H. (1996). Efficient tests for an autoregressive unit root. Econometrica, 64(4), 813–836.

    Article  Google Scholar 

  11. Fama, E. F., & French, K. R. (1987). Commodity futures prices: some evidence on forecast power, premiums and the theory of storage. Journal of Business, 60(1), 55–73.

    Article  Google Scholar 

  12. Government of India (2011). Annual Report, Ministry of Consumer Affairs, Food and Public Distribution, New Delhi.

  13. Granger, C. (1989). Forecasting in business and economics, 2nd Edition. San Diego: Academic Press.

    Google Scholar 

  14. Grudnitski, G., & Osburn, L. (1993). Forecasting S and P and gold futures prices: An application of neural networks. Journal of Futures Markets, 13(6), 631–643.

    Article  Google Scholar 

  15. Gulati, A., Chatterjee, T. & Hussain, S. (2017). Agricultural commodity futures: Searching for potential winners, ICRIER working paper 349.

  16. Hartzmark, M. L. (1991). Luck versus forecast ability: Determinants of trader performance in futures markets. Journal of Business, 64(1), 49–74.

    Article  Google Scholar 

  17. Henderson, B. J., Pearson, N. D., & Wang, L. (2014). New evidence on the financialization of commodity markets”. The Review of Financial Studies, 28(1), 1285–1311.

    Google Scholar 

  18. Jensen, M. C. (1978). Some anomalous evidence regarding market efficiency. Journal of Financial Economics, 6(2–3), 95–101.

    Article  Google Scholar 

  19. Kearns, J., & Manners, P. (2004). The profitability of speculators in currency futures markets, Working Paper, Reserve Bank of Australia.

  20. Kho, B. C. (1996). Time-varying risk premia, volatility, and technical trading rule profits: Evidence from foreign currency futures markets. Journal of Financial Economics, 41(2), 249–290.

    Article  Google Scholar 

  21. Konstantinidi, E., & Skiadopoulos, G. (2011). Are VIX futures prices predictable? An empirical investigation. International Journal of Forecasting, 27(2), 543–560.

    Article  Google Scholar 

  22. Konstantinidi, E., Skiadopoulos, G., & Tzagkaraki, A. (2008). Can the evolution of implied volatility be forecasted? evidence from European and US implied volatility indices. Journal of Banking and Finance, 32(11), 2401–2411.

    Article  Google Scholar 

  23. Kumar, R. (2017). Price discovery in some primary commodity markets in India. Economics Bulletin, 37(3), 1817–1829.

    Google Scholar 

  24. Kumar, B., & Pandey, A. (2011). International Linkages of the Indian commodity futures markets. Modern Economy, 2(3), 213–227.

    Article  Google Scholar 

  25. Ma, C. W. (1989). Forecasting efficiency of energy futures prices. Journal of Futures Markets, 9(5), 393–419.

    Article  Google Scholar 

  26. Malliaris, A. G., & Urrutia, J. L. (1996). Linkages between agricultural commodity futures contracts. Journal of Futures Markets, 16(5), 595–609.

    Article  Google Scholar 

  27. Malkiel, B. G., & Fama, E. F. (1970). Efficient capital markets: A review of theory and empirical work. The journal of Finance, 25(2), 383–417.

    Article  Google Scholar 

  28. Miffre, J. (2001a). Efficiency in the pricing of the FTSE 100 futures contracts. European Financial Management, 7(1), 9–22.

    Article  Google Scholar 

  29. Miffre, J. (2001b). Economic activity and time variation in expected futures returns. Economic Letters, 73(1), 73–79.

    Article  Google Scholar 

  30. Miffre, J. (2002). Economic significance of the predictable movements in futures returns. Economic Notes, 31(1), 125–142.

    Article  Google Scholar 

  31. NCDEX (2014). Market Data, Retrived September 1, 2014, from

  32. Schwarz, T. V., & Szakmary, A. C. (1994). Price discovery in petroleum markets: arbitrage, cointegration, and the time interval of analysis. The Journal of Futures Markets, 14(1), 147–167.

    Article  Google Scholar 

  33. Sehgal, S., Ahmad, W., & Deisting, F. (2014). An empirical examination of the process of information transmission in India’s agriculture futures markets. Journal of Quantitative Economics, 12(1), 96–125.

    Google Scholar 

  34. Shihabudheen, M. T., & Padhi, P. (2010). Price discovery and volatility spillover effect in indian commodity market. Indian Journal of Agricultural Economics, 65(1), 101–117.

    Google Scholar 

  35. Tsay, R. S. (2010). Analysis of financial time series. New Jersy: Wiley.

    Google Scholar 

  36. Wang, C. (2004). Futures trading activity and predictable foreign exchange market movements. Journal of Banking and Finance, 28(5), 1023–1041.

    Article  Google Scholar 

  37. Welch, I., & Goyal, A. (2008). A comprehensive look at the empirical performance of equity premium prediction. Review of Financial Studies, 21(4), 1455–1508.

    Article  Google Scholar 

  38. Yoo, J., & Maddala, G. S. (1991). Risk premia and price volatility in futures markets. Journal of Futures Markets, 11(2), 165–177.

    Article  Google Scholar 

Download references


I would like to express my sincerest gratitude to Prof Sunil Kanwar for his suggestions and immense help. A heartfelt thanks to Prof Chandan Sharma, Chandan Singha and Chhaya Chanchal for their feedback. My sincere thanks to the conference participants for their insightful comments at: i) 14th Annual Conference at ISI, New Delhi, December 19-21, 2018 ii) National Conference at CTRPFP (CSSS), Calcutta, March 15, 2019, and iii) 1st CMC at NISM, Mumbai, November 21-22, 2019.

Author information



Corresponding author

Correspondence to Raushan Kumar.



Architecture of Artificial Neural Network (ANN)

The neural network comprises three layers: the input layer, the (hidden) intermediate layer, and the output layer. While the input layer is analogous to the regressor variables in the traditional regression framework, and the output layer corresponds to the regressand, it is the (hidden) intermediate layer that is the distinguishing component of a neural network framework. This intermediate layer tends to get denser and, indeed, multi-layered as the phenomenon of interest (or the data-generating process as it were) gets more complicated.

An artificial neural network consists of processing elements, we call it neurons.Footnote 10 Each neuron is like a variable. The arrows linking the neurons signify the parameters of neurons (Fig. 1). The parameters of the neurons are the weights associated with them. In this technique, we connect the input to the output data via a system of neurons and these neurons learn from other neurons. The following are the steps used in constructing an artificial neural network.

Fig. 1

Artificial neural network

First, we decide the structure of the network which is explained in Fig. 1. It includes number of input nodes, number of hidden layers and nodes, the transfer function and the output node. The number of input nodes are our lagged explanatory variables such as the gram futures prices, the real rate of interest, the basis defined as the difference between spot and futures prices, and the futures price of wheat in the USA, ratio of high to low price. The first lag of the wheat futures prices is also an explanatory variable. Therefore, the number of input nodes are six. We have one hidden layer (three neurons), and one output node, i.e., the wheat futures prices. The output node is our dependent variable.

Second, we split the data on explanatory variables and dependent variables into two groups. The first group is used to train the network, and the second group is used to validate the network using out-of-sample forecast. Our sample size is 891. We have used the initial 800 sample points for fitting the network, and the remaining 91 sample points to validate the model.

Third, we scale all explanatory as well as output variables in the range of 0 to 1. It is required because of the mathematical structure of the network. The rescaling of the data has no impact on the pattern. We rescale the data using the following formula:

$$Scaled\, value = \frac{Actual - Minimum}{{Maximum - Minimum}}$$

where Minimum \(=\) Expected minimum value in data. Maximum \(=\) Expected maximum value in data.

Fourth, we set the initial training weight and begin a training epoch. An epoch implies the computation of errors and change of weights by processing the entire sample in the training data set. In addition, we give initial values to all the weights in the network. The initial weight affects the final solution. The good starting point of the initial weight is unknown. Most programs recommend the randomization of the training weights between − 1 and 1 (DeLurgio 1998). We have used RATS software, and it determines the initial weight randomly between − 1 and 1.

Fifth, we provide the scaled inputs at neurons 0, 1, 2, 3, 4 and 5 (Fig. 1), and we designate these neurons as \(I_{0}\), \(I_{1}\), \(I_{2}\), \(I_{3}\), \(I_{4}\) and \(I_{5}\), respectively. Generally, we name input node as \(I_{j}\) and output node as \(O_{j}\). At input layer, there is no transformation of explanatory variables, so output of an input neuron is the same as its input value.

Sixth, every neuron in the input layer receives its scaled value and transmits it to every neuron in the hidden layer. These scaled inputs are weighted and provide input to hidden neurons in step (7). There is parallel processing of inputs at other two hidden neurons. In our network, there is one hidden layer with three neurons.

Seven, we give weights and sum inputs to receiving nodes.Footnote 11 This implies that at every hidden neuron, we give weights to the output of input neurons and sum it. Therefore, the input to neuron 6 is written as:

$$I_{6} = W_{06} O_{0} + W_{16} O_{1} + W_{26} O_{2} + W_{36} O_{3} + W_{46} O_{4} + W_{56} O_{5}$$

We determine the weights and values after several iterations of the training data. The weights are memory and intelligencef the network. They are similar to the coefficients of the regression model. But, weights in the neural network are hard to interpret.

Eight, we transform the weighted inputs at each hidden node to outputs ranged from 0 to 1. We employ the logistic function to express the relationship between input and output of neuron. This is a nonlinear transformation.Footnote 12 Output at neuron 6 is

$$O_{6} = \frac{1}{{\left( {1 + e^{{ - I_{6} }} } \right)}}$$

where \(e\) is the natural number whose value is 2.718…

The output value of \(O_{6}\) is one input for neuron \(I_{9}\) (Fig. 1).

Nine, we give weights and add hidden neuron outputs as input to output node. The output neuron is the final neuron. From Fig. 1

$$I_{9} = W_{69 } O_{6} + W_{79 } O_{7} + W_{89} O_{8} + \theta_{3}$$

where \(\theta_{3}\) is the bias of the forecast. It is like a constant term in regression equation.

Ten, we transform the weighted inputs \(I_{9}\) to output of \(O_{9}\) ranged from 0 to 1. \(O_{9}\) is the final output of the neural network. Again we employ the logistic function. Output at neuron 9 is

$$O_{9} = \frac{1}{{\left( {1 + e^{{ - I_{9} }} } \right)}}$$

Eleven, we then calculate errors in the output. We compare the scaled ANN output value \(O_{9}\) to the true output (scaled) \(D_{9}\), and compute the error.

$$e_{i} = D_{9} - O_{9}$$

where \(i\) denotes the observation number in the training data.

Twelve, in this step, there is back propagation of errors, i.e., there is adjustment of weights in the training so as to achieve the minimum residual mean standard error (RMS) value (DeLurgio 1998). The method of spreading information about errors from the output layer back to hidden layer is termed as back propagation.

Thirteen, we continue the epoch by repeating the steps from fifth to twelfth for all input data.

Fourteen, there is calculation of epoch RMS value. If we get the low value of the RMS,Footnote 13 we move to the next step, otherwise we repeat the steps (5) to (14).

Fifteen, we validate the ANN results with out-of-sample data. The model is valid if the training RMS is consistent with the out-of-sample RMS.

Sixteen, we then use the model for forecasting.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Kumar, R. Predicting Wheat Futures Prices in India. Asia-Pac Financ Markets 28, 121–140 (2021).

Download citation


  • Futures market
  • Forecasting
  • Artificial neural network (ANN) and agricultural derivative market

JEL Classification

  • G13
  • C45
  • Q14