# Introduction ommodity prices have been very volatile in the most recent years, particularly grains, they reached an exceptional peak in the year 2008, and then they declined sharply, but started rising again in 2010.Volatility must be distinguished from variability, volatility is a measure for scale and speed of prices evolution, it includes variability and uncertainty, and it describes prices evolution that we cannot expect using forecasting models, it refers to the unpredictable changes in prices. Whereas variability refers to changes in prices due to the variation of fundamentals, it can be predictable. It has commonly admitted that the mid-2000s marked the start of a trend of steeply rising commodity prices, accompanied by increasing volatility. This period was characterized principally by an increased demand, in particular, in emerging economies (China, India, Brazil?etc), and the use of cereals in the production of bio fuels, at the same time, supply was declined sensibly as a result of the adverse effects of climate change and a decline in the productivity of agricultural lands. Although this volatility cannot be explained only by these factors, another major factor is the phenomenon of financialization of commodity markets, volumes of financial investments in commodity derivatives markets has increased significantly since 2004. In fact, producers have been very risk averse in this situation, and they find in future markets the mean to hedge their positions against uncertainty that lead to sharply prices changes. From the other side, investors have been engaging in commodities markets for diversification ever since it becomes evident that commodity futures contracts exhibited the same average returns as investments in equities, while over the business cycle their returns were negatively correlated with those on equities and bonds. The attractiveness of commodity futures contracts also relates to the good hedging properties against inflation. All these changes in the ten recent years led to the increasing role of the financial motives, financial markets and financial actors in the operation of commodity markets, hence the increased financialization of agricultural commodity markets. Many researchers investigate the relationship between speculation activity and commodity prices volatility, while some researchers support this linkage, others do not support it for different reasons. The purpose of this paper is to identify the impact of financial speculation on commodity prices volatility through the behavior of commercials (producers) about risk. # II. # Speculation: Liquidity Versus Volatility Financialization is the phenomenon which characterized the agricultural future markets since 2000, from 2003 to 2008 funds allocated to commodity index replication trading strategies have grown from 15 billion dollars to 320 billion dollars, at the same time, the prices for the 25 commodities that make up these indices have risen by an average of 200%. In fact, speculation has been raised excessively in commodity future markets, and its impact has been hotly discussed by researchers in recent years, most of them think that the volatility which characterized commodities future markets is a consequence of excessive speculation; Hedge fund manager M.W. Masters is the most ardent supporter of the speculation impact on commodity prices volatility; he argues that unprecedented buying pressure from index investors created a massive bubble in commodity futures prices, and this bubble was transmitted to spot prices * They concluded that significant causality exists between assets under management variability of commodity funds and prices variability, but mainly from commodity index funds. However, no significant , so price spikes were driven in large part by a new type of speculators in commodity futures markets. It means that changes in futures prices lead changes in spot prices more often than the reverse, as noted by M. Hernandez and M.Torero. Other researchers like Irwin and Sanders, despite their antagonism towards speculation impact, use a shorthand label for this argument as "Masters Hypothesis" to describe excessive price volatility induced by financial speculation. Ke Tang and Wei Xiong, in their work file titled "index investment and the financialization of commodities," found that commodities in the S&P GSCI and DJ-UBSCI had significantly greater volatility increases than did off-index commodities in 2008. So commodities price changes do not reflect only fundamentals changes, they argue that concurrent with the rapid growth of index investment in commodity markets, prices of non-energy commodities have become increasingly inter-correlated, and also correlated with Oil prices. This situation is a result of the speculation process started in 2000, it reflects the financialization of the commodity markets and helps to explain the large increase in the price volatility of nonenergy commodities around 2008.Hence, the price of an individual commodity is no longer determined solely by its supply and demand. Instead, prices are also determined by the aggregate risk appetite for financial assets, and the investment behavior of diversified commodity index investors. J. Cordier and A. Gohin (2012) in their analysis have been looking for an impact of speculation on cereals prices by analyzing the relationship, first, between assets under management of the commodity funds and the agricultural futures prices; second; they searched a sequential relationship between these variables through the commitment of commodity funds on related futures markets. causality was detected of commodity funds commitments on futures markets, they argue that this absence of causality is due to the ability of commodity funds to hedge their prices risk on the OTC market as a complement to the futures markets. On June 24, 2009, a report about excessive speculation in the wheat market was presented in the US Senate by C. Levin and T. Coburn; this report unveiled some key data that confirm the impact of speculation on commodities prices volatility, particularly, in the wheat market: "The amount of speculation in the wheat market due to sales of commodity index instruments has, correspondingly, grown significantly over the past five years. CFTC As notified by the FAO in the treaty of Rome (23 Juin 2010), Large commodity funds now hold about 25-35 percent of all agriculture futures contracts and, with other investors, have become an important source of liquidity to the market Futures contracts involving the formal obligation to sell or buy a given amount of a commodity at a specified time and price. They thus provide farmers and traders with an important defense » Having realized this, the US Senate voted in 2010 the Dodd-Frank Act in order to limit speculation in commodity markets, this law has faced some critics believing that the act will ultimately hurt economic growth, like limitation of the bond market-making role that banks have traditionally undertaken, this situation, in turn, can lead to lessen market liquidity. Researchers like S. Irwin, S. Sanders, Gilbert, Stoll and Wally, Hamilton and Wu, consider that speculation activity is source of liquidity in agricultural commodity market, and, based on normal backwardation theory, they think also that it is a condition sine qua non to reach equilibrium between spot and future prices in these markets, thus, they do not support the Master's hypothesis. Gilbert has used time-series test, such Granger causality test to analyze the impact of speculation on cereal prices; findings report that there is no significant time-series relationship between weekly financial index trading and returns in wheat, corn, and soybeans markets. or "hedge" against price risks. However, it is very important to note that only two percent of futures contracts end in the delivery of the physical commodity as they are traded, generally, before their expiration date. As a result, such contracts, or obligations, are drawing growing numbers of financial speculators and investors, especially as they can provide attractive returns when equities and bonds may become unappealing * Irwin and Sanders think that bubble argument does not withstand close scrutiny, and excessive speculation is not an argument for the volatility of agricultural commodities prices volatility for four reasons . * -The impact should be more evident in a shorter time. Nonetheless, the CFTC provides only weekly data about financial index positions in agricultural futures markets, and for this reason, the impact of changes in index positions will be less clear, hence reducing the power of time series methods to detect its impact. It's well known that the role of information flows is crucial for prices formation, the EMH (efficient market hypothesis) postulates that all publically available information is immediately reflected in prices, even private information available only to individual market participants is reflected in the price through the effects of the transactions of the persons in possession of the information, for this reason, commodity price developments would reflect nothing but information on fundamentals. However, market participants make trading decisions based on factors that are totally unrelated to the perspective commodity, such as portfolio considerations, or they may be following a trend, ignoring changes in fundamentals. Thus, the trading decision process is characterized by considerable uncertainty, particularly in agricultural markets, most of the traders follow other participants in trading decisions, witch leads to creating the so-called "intentional herding," and this behavior is accused of creating a speculative bubble that cannot be justified by changes in fundamentals. # III. Herd behavior in Agricultural Future Markets Market participants continuously update their expectations about prices evolution from the inflowing public and private information. As a result, prices move upward or downward when new information is publicly available or when private information leads to transactions that affect prices. It means that market participants evaluate their assets based on fundamentals, that is what we call an act fully rationally, but when they ignore their own information and variations in fundamentals to follow other market participant's decision, market efficiency will not be reliable, and prices evolution cannot be explained solely by fundamentals variation. In fact, traders can engage in herd behavior in some circumstances, particularly when the market is characterized by a big uncertainty, this behavior consists to mimic the action of a dominate group of investors, it can be qualified as an irrational behavior as it may also be fully rational. For example, an investor who is ready to invest in the securities of an issuer, ignoring other market participant's decision, but he changes opinion when he realizes that other investors have decided to abandon. Some recent models consider that the herd behavior is a deviation of rationality, this behavior is known as a "noise trading," it means that traders decisions are affected by a pseudo-signals, some market participants take a sell or buy decision only to assign supply and demand, which lead to affect prices. Noise trading can be also described by changes in beliefs and sentiments. As a result, traders can, for example, take decisions based on an algorithmically software independently of any changes in fundamentals, like selling after prices fall, and buying after prices rise. # Global Journal of Management and Business Research Volume XVIII Issue I Version I # Year ( ) Herd behavior can be rational, in this context spurious herding must be distinguished from the intentional herding, as it described by Bikhchandani and Sharma, this behavior consists to take the same decision unintentionally when traders face the same circumstances independently from the other market participants decision. This behavior does not contradict the EMH, for example, banking panics. Unlike the previous, intentional herding may be described by following other market participant's decision because of a psychological impact, and they behave so for four motives: -Imitation that arises when traders and their employers doubt their own abilities to take a right decision. -When agents invest on behalf of others, herding can be a result of a compensation incentive; Thus, they align their positions with benchmark portfolios. -Conformity-based herding relates to an alleged intrinsic performance of individuals for conformity. -Imitation based on believing that market participants can glean information by observing the behavior of other agents. In spite of this distinction between various herding types, if all these acts lead to affect price movements, early moves will benefit the most. Imitation by followers will gradually become less profitable the larger it is delayed, and the greater becomes the probability that newly arriving public information will alter the informational cascade, thus, motivation to engage in herding behavior decline progressively until it ended, and the extent to which herding affects prices depend on the degree of uncertainty. Within that period, it will be difficult to distinguish the well informed from the uninformed agents, called the followers. In this situation, market participants may believe mistakenly that most agents possess accurate information, hence the dramatic effects on prices that can lead to bubbles and excessive volatility because of the ensuing confusion, which allows the uninformative herd behavior to affect drastically prices. This analysis shows that market participants can react for some reasons, whether they are rational or irrational their behavior can instantaneously push prices to deviate from fundamentals for a long period creating a big uncertainty. Therefore the decision process became more complicated for a risk-averse agents, in particular, producers and customers, this effect was more obvious in cereal market in 2007 until 2012. It has become very difficult to predict and analyze agent's behavior, empirical work files realized cannot sufficiently provide evidence about this phenomenon, and some findings are in favor others against of the presence of this herd behavior and its impact on prices. It is for this reason that we conclude that prices movements depend in general on fundamentals changes, and financial investor's game in the market (spoofing * , layering * According to normal backwardation theory ?etc). # * IV. # The Decision Under Risk and Uncertainty , the difference between the forward prices and the expectation of spot prices can be justified by a speculator remuneration called risk premium, this remuneration can change proportionately to the degree of risk aversion that can be different from an agent to another. In this situation, it can be more evident, under uncertainty, ensuing by a herd behavior, that we can expect an indirect but significant impact of speculation on prices through excessive risk aversion of producers and customers. Act in a situation where the information is available and symmetrically distributed is not a problem for the various market participants, because the ensuing price would be right, it is an equilibrium price. However, if the market is characterized by great volatility (described by variability and uncertainty), the ensuing price may not reflect supply and demand tendency, and the future price cannot be explained based on a future spot price expectation. Therefore market cannot regulate itself. The economic theory developed in XIX century was static. It assumed that information is perfectly and symmetrically distributed, and this was not the case for the cereal market in the last decade, risk and uncertainty were a principal characteristic of the market that results from the various wrong market signals due to strong speculation and blind herd behavior. Consequently, * An illegal practice, it is also a form of market manipulation in which investors use visible non-bona fide orders to deceive other traders as to the true levels of supply and demand. * Layering is a form or variant of spoofing where the trader places several orders a few ticks apart to give the appearance of buying or selling, which cause the midpoint of the spread to move away from those orders, and the same trader executes a trade on the opposite side of the market. * Developed by J.M Keynes, based on this theory, a market is said to be in contango when future prices lie above spot prices, and it said in backwardation when the future prices are below the expected future spot prices. This theory is used to explain the relationship between the future prices and the expected value of the spot prices of the commodity at some future date. Normal backwardation suggests that the future prices will be bid down to a level below the expected spot price, and will rise over the life of the contract until the maturity date. On the maturity date, future prices are equal to spot price. , if it comes up tails the second time, the prize would be ,and if it comes up tails the n time, the prize would be n 2 MU. Knowing that probability of a consequence of n flips is: 1 n 2 , the expected value of the game( E(x)) is the sum of the expected payoffs of all the consequences; If it refers to mathematical analysis, taking into account mathematical expectation as it is justified by Pascal and Fermat, this game may not contain any contradictions. However, the expected value of the game is an infinite number of dollars, which lead us to believe that the game organizer cannot reward the winner if E(x)tend towards , he should have established a higher price for the lottery. And from the other side, the rational gambler would not accept to pay even 100 MU, for example, to enter such a game knowing that the prize could be only 2 MU. Then something has gone wrong with this way of thinking about the game, which has become, following this logic, not playable. This paradox has questioned the concept of mathematical expectation. * Monetary unit D. Bernoulli claimed that two analysis criteria ignored in the previous analysis: -Behavior and individual characteristics. -The evaluation method of the results, which calculated, based not on monetary units, but on utility-based units. The utility theory postulates that people behave as if they make a decision by assigning imaginary utility values to the original monetary values, and knowing that any agent reaches a saturation point for utility. There is a decline in the marginal utility that person derives from consuming each additional unit of any product, and the saturation level may differ from agent to another. Thus, someone may be interested in a prize of 100 MU, but the same prize cannot be interesting for another agent, and there is no gambler who can continue to play until E(x)tend towards . D. Bernoulli argues that any slow increase of wealth ( w ), the increasing in utility ( u ) is given by: w u w dw du w w u ln 1 1 For Bernoulli this hypothesis is valid for a most of agents, hence in St Petersburg game, the mathematical expectation is becoming a moral expectation, and this does not tend to infinity, but to a finite number: This means that, when n (number of flips) tend towards infinity, the moral expectation may tend to a finite number. Thus, there is no gambler who can continue to risk until infinity. This idea was carried forward later in 1944 by E. Borel, J.V. Neumann and O. Morgenstern, concretized in a theory of games and economic behavior, based principally on realistic hypothesis, particularly uncertainty, asymmetrical information and the probability of results. Global # b) Expected utility theory (VNM) According to VNM analysis, if economic agents evaluate results based on their utility, not by a monetary unit, the situation of uncertainty can be described as follows: Let E be the finite set of possible events, and P a set of the probability distribution on this set ) ( 1 r u , 1 p ),( ) ( 2 r u , 2 p )???..( ) ( n r u , n p )]( Considering possible outcomes as a wealth (w) of an economic agent, we obtain the following formula: [( ) ( 1 w u , 1 p ),( ) ( 2 w u , 2 p )???..( ) ( n w u , n p )] VNM argue that, economic agents choose, in an uncertain situation, based on an expected utility carried from every situation as follows: ) ( ) ( ) , .( ),........ , ( ), , ( 1 1 2 2 1 1 w EU w u p w p p w p w p w U n i i i i n i i i i n n This equation represents the formula that describes the expected utility of an economic agent. Thus, individual faces a preference of decision-making in an uncertain situation will always prefer actions that maximize expected utility by comparing -Individual who prefers ) ( f w to E(wf) ) ( ) ( f f w EU w UE This behavior is considered as a risk-seeking. Hence the individual utility function must be represented by a positive exponential function w e w U ) ( , for example. -The third type of behavior is the indifference, or risk neutral, ) ( ) ( f f w EU w UE , which can be represented by a linear function ( b aw w U ) ( ), for example. Indeed, D. Bernoulli has explained one type of behavior; it is a risk aversion behavior, represented by the logarithmic function. We will go further to consolidate our ideas and hypothesis, it consists to describe a commercial (producer) behavior in cereal market; this commercial (producer) is supposed risk averse under uncertainty in relationship simultaneously to a fundamentals changes and to the wrong market signals as a result of a great speculation, as it described above. Based on VNM deduction, the utility function that describes the commercial behavior is taken as lnw, this function can be introduced in our econometric model to seek the impact of speculation on cereal prices through commercials (producers) behavior. ) see afterwards the improvement of the estimation due to the lags' values of the variable X taken into account. Money managers, and other reportable) on the behavior of commercials, but not a direct impact on cereal prices, using VAR model. Commercials are supposed, in this study, risk-averse as long as they use hedging instruments, and they engage in herd behavior. # V. Empirical Analysis of Wheat Prices Volatility # a) Data Description CFTC is an institution whose mission is to regulate, control and collect information, it aims to protect market users and their funds, consumers, and the public from fraud, therefore, it provides information in periodic reports about the commitment of traders, these reports are available in both a short and long format. The supplemental reports show aggregate futures and options positions of non commercials, commercials and index traders in 12 selected agricultural commodities. Statistical data used in this study is gathered from the Cbot market. Concerning traders position; the data is collected from weekly reports of the CFTC, monthly wheat and corn prices are available in UNCTAD and FAO web site, prices are expressed in dollars per ton. A chosen time series are used from June 2006 until December 2015, the study period contains 115 observations. Using this data, we proceed to estimate the time series data using the ninth version of Eviews software. # b) Model Specification In this study, it is a question of regressing historical price values on actual prices, and on the other variables that may have a significant impact on future prices evolution, the other variables taken in account in our model are: the speculation position variation (long and short position) and the spread as follows: t t t t spl sal lal t f t f ) ( ) ( ) ( )) 1 ( ( ) ( 4 3 2 1 0 ( ) (t f ) : As utility function of a professional at the time t , such as f(t) = U(x), x represents the wealth of the professional and the price of a ton of wheat. Considering a risquophobe commercial (as was our hypothesis), his utility function can be as, U(x) = ln x. # Let x dx x d t f x t f 1 ln ) ( ln ) ( , such as, x represents the wealth of the commercial and it is considered as the price of a ton of wheat. Before estimating the model, all variables should submit the various stationary tests, and detect if any seasonal effect exists. # c) Stationary tests A time series stationary means that its variance and expectation are independent of time variation. Otherwise, we consider the time series as nonstationary. Thus, we cannot estimate an econometric model that its variables are not stationary, because the impact of explanatory variables on explained variables would be confused by the time variation. A common test used is the ADF test (Augmented Dicky fuller test), based on three types of models: -The first one does not contains any constant or time drift, this model is written: t n i i t i t t y X X 1 1 1 -The second is a model with constant and time drift, this model is written: t n i i t i t t y X X 1 1 1 1 -The third model contain a constant, but not a time drift, it is written : t n i i t i t t y X t X 1 1 1 1 According to ADF test, if H0is selected inany model of three models, we qualify the process as nonstationary, in this situation; the estimated value of t of student associated to parameter exceeds the critical tabulated value of Mackinnon (ADF tab): It means that: H0 : = 0 H1 : < 0 We accept H0, and we reject H1 if ADF cal>ADF tab. Otherwise, we accept H1 and we reject H0. Based on this table realized from results obtained from Eviews9 software, we have noticed that ADF cal>ADF tab for each model, therefore we accept HO and we reject H1, it means that the first, second and the third model have at least a unit root, so, we judge the series Ux as non-stationary, it is a kind of DS (differency stationary). Similarly, as for the first variable, we proceed for the other variables, and we conclude that the same results and analysis are obtained. This means that all time series are not stationary for all variables at a critical level of 5%. As the variables are not stationary at a critical level of 5%, we proceed with another alternative approach to make them stationary; this approach consists in testing the stationary of the first difference of the model. The obtained results are presented in the following table : 1 st This table indicates that ADF cal5%), we will then select the variables of the model by eliminating variables with a probability superior to , after that we should proceed to the reestimating of our model as follows: Based on this table and Dicky-fuller test, we conclude that residuals are stationary, we can then estimate our model in the following step: # e) Estimation of the Mode We proceed in what follows to the estimation of our model using Eviews 9 software in order to describe the relationship between risquophobe behavior of commercials (professionals) and the past values of wheat prices and speculators positions in the long run. From this table we not that the observed R²= 17.24%>?, so we cannot reject the null hypothesis, this means that there is no Heteroskedasticity in residuals series. -The third test that we must check too, is the normal distribution of residuals, for that purpose we can use a Jarque-Bera statistics test as follows: -We confirm that the probability is superior to ?, so we accept the null hypothesis and we reject the alternative hypothesis, it means that residuals are normally distributed. # g) Economic interpretation of the obtained results The obtained results confirm our theory about the significant impact of speculative positions on wheat prices volatility through commercials (professionals) behavior. Hence, our econometric model can be written as: L(Ux)b = 0.972169Lpx+ Knowing that all variables are borderline I(1), this means that all independent variables (past values of prices, speculator long and short positions) have a significant impact on commercials risk aversion with a single period lag ( one month). Passed values of wheat prices are integrated into our econometric model with a positive sign, and a coefficient = 0.97, it indicates that the fact that commercials are very sensitive to prices evolution, and that is how it should have been, their decision to buy or to sell depend on the future price development, based on passed development process. Speculative long positions are integrated with a positive sign and a coefficient = 8.40, it indicates that the long position of speculators has an important effect on utility function of commercials, thus a positive effect on their risk aversion. Therefore, any long position variation of speculators in future market can create a herd behavior wave, which stimulates the emergence of a new speculative buying wave in the commodities market, conducting to a massive increase of prices, because the market will transmit a spurious positive signal of buying. Speculative short positions have a lower impact (coefficient = 4.78), but they are integrated with a negative sign. Consequently, the impact will be negative on commercials behavior, it means that the fact that speculators get rid of their buying positions, commercials risk aversion increases, which will affect negatively the utility function, conducting to a reticence vis a vis to buying decisions, which stimulate a sharp drop of wheat prices. # VI. # Conclusion The obtained results indicate that commercials facing financialization of commodities markets have become uncertain, because of strong speculation, and a phenomenon of herd behavior, on that point, the use of hedging techniques is a valuable argument of commercial risk aversion. Several research studies indicate that there is no impact of speculation on prices volatility, particularly in cereal markets, however, the use of future market instruments justify the uncertainty and the risk aversion of commercials, resulting from a big wave of speculation accompanied with a herd behavior, which can stimulate, for its part, the soaring or the sharp drop of prices. We have tried to analyze the commercials behavior in cereal market based on a VNM expected utility theory, and we concluded that the impact of speculator position variation is evident in the long run for the wheat prices, the fact that the commercials behavior is affected. As a result, our theory based on the possible effect of commercials risk aversion, which is subject to the speculation impact, on prices volatility is well verified through this econometric modelling. ![words, individuals make decision by comparing mathematical expectation of possible outcomes utility, and the utility of every possible outcome: UE(W)~EU(W). Therefore, three types of economic agents can be distinguished: Individual who prefers E(wf) to](image-2.png "-") ![VAR modelling was introduced for the first time by Sims in 1980 to mitigate the failures and deficiencies of macro-econometric models, because of their incapacity to forecast economic crisis in1973, 1979. In the same context, Granger developed a new concept of causal link well known by Granger causality. This consist plainly in demonstration if the variable X cause (in the sense of Granger) a variable Y through random shocks, looking first the extent to which past values of the variable Y explain the actual value of the variable Y, and Global Journal of Management and Business Research Volume XVIII Issue I Version I Impact of Financial Speculation on Commodities Prices' Volatility through Commercials Risk Aversion "Application for Wheat Prices"](image-3.png "") ![](image-4.png "(") ![As a speculator long position variation (swap dealers, money managers, and other reportables) for the period t.](image-5.png "") ![a speculator short position variation (swap dealers, money managers, and other reportables) for the period t.](image-6.png "") ![a spread (swap dealers, money managers, other reportables) for the period t.](image-7.png "") ![Impact of Financial Speculation on Commodities Prices' Volatility through Commercials Risk Aversion "Application for Wheat Prices"](image-8.png "B") 2nmonetary units. For example, if the coin comes up tailsthe first pitched, the prize would be1 2 MU *2 24MU EU) (wE(ln) w1/2ln21/4ln41 .......2nln2n12n1ln2n112n1ln2n1138 .1YearVolume XVIII Issue I Version I( ) e 1e , 2n e .. ..........as possible events, andr 1r , 2n r .. ..........considered as the results assigned to each event,1 p, 2 p.. ..........pn/i p1consideredasprobabilities associated to each event which lead to aresult r.The set of combinations [( 1 r , 1 p ),( 2 r , 2 p???( n r , n p )] describes an uncertain position whereplenty of events are possible. However, if we refer toBernoulli's analysis, we may introduce the utility criteria,and this situation should be described as:[ VariableCoefficientStdErrorTstatProbaCritical value (at 5%)TcalModèl 1Ux(-1)-0.0048620.007525-0.646170.5195-1.94368-0.64617Modèl 2Ux(-1)C Trend-0.087123 -0.066158 25.199230.034330 0.058346 9.772427-2.53783 -1.13387 2.5786050.0126 0.2593 0.0113-3.45007-2.53783Modèl 3Ux(-1)C20.04007 -0.0817848.659792 0.0340492.314152 -2.401920.0225 0.0180-2.88719-2.40192Source: Authors' estimations * NM.Aulerich, S.Irwin, P.Garcia, "Bubbles, food prices, and speculation: evidence from the CFTC's daily large trader data files", October 2012, P2. © 2018 Global Journals Impact of Financial Speculation on Commodities Prices' Volatility through Commercials Risk Aversion "Application for Wheat Prices" * «Agrestconjoncture Céréales 2014/229, Janvier 2014 * Excessive speculation in the wheat market CLevin TCoburn 24 june 2009 United states Senate * La réduction du risque à l'aide des marchés à terme, Gestion des risques en matière de revenu dans le secteur agricole JCordier 2000 OCDE * Cadre conceptuel de la gestion du risque agricole, les types de risque et les instruments de gestion JCordier 2006. 10 octobre 2006 Paris Intervention au Colloque du COPEIAA * Irwin « the impact of index funds in commodity futures markets: a systems approach », the journal of alternative investments DSanders S summer 2011 * Examining the dynamic relationship between spot and future prices of agricultural commodities MHernandez MTorero June 2010 IFPRI discussion paper 00988 * Bubbles, food prices, and speculation: evidence from the CFTC's daily large traders data files NMAulrich SHIrwin PhilipGarcia October 2012 * index investing and the financialization of commodities" working paper, department of economics TKe WXiong 2010 Princeton University * Revu Tema n°1, « Tension sur les marchés des céréales »,p35,36, premier trimestre 2007 Paris * La lettre des professionnels « les marchés mondiaux des matières premières agricoles », numéro spécial Octobre 2010 Rapport du CIC * Rapport de la FAO, « perspectives alimentaires 2014 * « perspectives économiques et sociales, la volatilité des prix sur les marchés agricoles », décembre 2010. 13. Communiqué de Momagri « agricultural marketvolatility Paris 2010 Rapport de la FAO * Jaques Carle « les enjeux agricoles mondiaux et la volatilité des marchés » mars MomagriCommuniqué De 2012