Medical, Pharma, Engineering, Science, Technology and Business

^{1}Department of Agricultural Economics, University of Gezira, Wad Medani, Sudan

^{2}Department of Economics, University of AL-Butana, Rufaa, Sudan

- *Corresponding Author:
- Mohammed OA Bushara

Department of Agricultural Economics

University of Gezira, Wad Medani, Sudan

**Tel:**(+249)511 - 841623

**E-mail:**[email protected]

**Received date** January 04, 2016; **Accepted date** January 27, 2016; **Published date** January 30,
2016

**Citation:**Bushara MOA, Abdelmahmod MKA (2016) Efficiency of Selected
Sudanese Cattle Markets: A Bivariate Cointegration Approach (1995-2011). Int J
Econ Manag Sci 5:321. doi:10.4172/2162-6359.1000321

**Copyright:** © 2016 Bushara MOA, et al. This is an open-access article distributed
under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the
original author and source are credited.

**Visit for more related articles at** International Journal of Economics & Management Sciences

The importance of the livestock trade to the national economy of Sudan is significant; however, Sudan was adversely affected by the global crisis through a decline in oil and other external receipts. The main objectives of this study is to investigates price movements among important livestock markets in the Sudan to explore their performance and pricing efficiency. The stationarity of data tested using the unit root test and then market integration was tested using bivariate cointegration analysis. The study found strong evidence of cointegration of pairs of markets. The error correction mechanism adjusts significantly to shocks to its equilibrium relationship. The estimated coefficients for ECM were fluctuated between 4% and 24% and significant at 1% level, suggesting slowing adjustment towards the long-run equilibrium. This implies that any shock that forces prices from their long-run value would take a long time for prices to return to their equilibrium, although the speed of adjustment was highest in case of Omdurman on Nyala and lowest in case of Nyala on Elobied, this might be due to supply and demand relation between Nyala and Omdurman in sense that Nyala and Elobied were supply markets. The recent paved roads that linked between the markets might accelerate and facilitated the movement between these markets, in addition to a huge capital that allocated to livestock business recently.

**Market efficiency**; Cattle; **Coinegration**; **Bivariate**; Sudan

Most researchers agree that the problems of **livestock **marketing in
Sudan are limited to the specific problems which can be summarized in
a weak infrastructure especially in the area of transport and veterinary
services, lack of finance, areas of production distant from areas of
consumption and together with lack of suitable transport render
animals weak and meat quality low, smuggling especially across the
borders to Egypt, Eritrea and Libya Markets that are not integrated
may convey inaccurate **price **signal that might distort producers
marketing decisions and contribute to inefficient product movement
[1] and traders may exploit the market and benefit at the cost of
producers and consumers.

Testing for the existence of cointegration among economic variables
has been widely used in the empirical literature to study economic
interrelationships. Its existence would imply that the two series would
never drift too far apart. A non-stationary variable, by definition,
tends to wander extensively over time, but a pair of nonstationary
variables may have the property that a particular linear combination
would keep them together, that is, they do not drift too far apart.
Under this scenario, the two variables are said to be co integrated, or
possess a long-run (equilibrium) relationship. Several studies of market
integration have been done in previous literature. Concepts of market
integration and market efficiency present cornerstones of modern
economics. Yet, the discipline struggles with the important, practical
challenges of clearly defining a market empirically and of establishing
whether markets are efficient in allocating scarce goods and services
[2]. Babiker and Abdalla [3] studied price movements among
important sheep markets in the Sudan to explore their performance
and pricing efficiency. Six geographically separated livestock wholesale
markets were tested spatially, using [4] Cointegration test and timeseries
price data for the period 1990-2004. Spatial analysis of the whole
dataset indicates the absence of cointegration among the selected
markets, while a subset of the data, for the period 2000-2004, after
some infrastructural facilities were introduced, shows that the same
markets are co integrated. Ibrahim [5] concluded the pastoral **economy **as exists in the Sudan was still having strong potential despite the many
problem related to the overstocking, low levels off take and quality of animals breed, which had reduced productive capacities. These
problems resulted from long run neglect of this sector. Idris [6] studied
the livestock marketing with the reference to Southern Darfur Region.
He categorized the livestock markets according to supply of animals
and their location with regard to availability of transportation facilities
to primary, secondary and terminal markets. He also traced the cost
incurred in trekking cattle from Nyala to Khartoum. In his study he
also found that livestock **trade **was controlled by a small number of
merchants and the barriers to entry in this system were high. M. SA [7]
studied livestock farming systems and argued that several successive
years of low rainfall; decorticating and inappropriate usage of land led
to damage of some region lands and increased pressure on less affected
areas. He concludes that to maintain the balance between animals
and pasture in short run the off take percentage to the market should
be increased and pasture regeneration and investment in this sector
were essential steps on the long run. Further insights on livestock
markets cointegration were documented by El Agip [8] who examined
cointegration and causality in five livestock markets (Omdurman,
Medani, Elobied, Sennar and Nyala towns) using monthly nominal
prices from January 1990 through December 1999. He found that
spatial market cointegration was present between these cattle and
sheep markets and the leading price discovery location was Elobied, in
other words the system was supply driven. Babiker [9] studied market
price integration for livestock in Omdurman, Medani, Nyala and
Elobied. She applied several approaches which were Engle and Granger
bivariate cointegration approach, Granger causality test and Johansen
Multivariate approach to the nominal monthly prices of cattle and sheep for the period 1980m1-200412. She concluded that markets
were co integrated, and the system was centered on Omdurman which
means the market was demand driven.

The study focused on scrutinizing the cattle markets in Sudan by considering the prices of five livestock markets which were Elobied, Omdurman, Medani, Sennar and Nyala. The study covered the periods from January 1995 to December 2011. The data used in this study were monthly price which have been collected from the animal resources company, these prices were wholesale prices i.e. the selling price of a head of animal measured in Sudanese Pound (SDG). To attain the cointegration analysis the data should be in real terms to avoid spurious regression, so all price series were deflated by GDP deflator(base year 1994) rather than consumer price index. The deflated prices data were transformed in term of natural logarithm so as to attain a constant variance in the series, and then logged and deflated prices data were used in the empirical analysis.

Testing for cointegration at the first step requires testing the order of stationarity of the variables. Integration tests or unit root tests are a prerequisite for cointegration tests, thus, an econometric model cannot be specified unless its order of integration of the variables is known. The order of integration in the time series checked by the Augmented Dickey-Fuller [10] (ADF) and Phillips and Perron tests [11], which are the most widely used methods for unit root tests. According to Babiker, in testing cointegration two conditions must be fulfilled: first the data series must have similar statistical properties; in particular, they must be integrated of the same order, because a variable with a constant mean cannot explain movements in a variable whose mean is changing through time. The second condition for cointegration is that there should be some linear combination between the data series. If and only if the hypothesis of no cointegration is rejected an error correction model (ECM) would be estimated to integrate the dynamics of short run (changes) with long run (levels) adjustment process.

The Engle-Granger residual-based test for cointegration is simply unit root test applied to the residuals obtained from OLS estimation of the following Equation:

Y_{t1} = α + βY_{t2} + u_{t} (1)

Where Y_{t1} and Y_{t2} are the two price series and ut is the error term.
This model called cointegration regression.

Under the assumption that the series are not cointegrated, all linear combinations including the residuals from OLS, are unit root nonstationary. Therefore, a test of the null hypothesis of no cointegration against the alternative of cointegration corresponds to a unit root test of the null of nonstationary against the alternative of stationarity. Accordingly, Engle-Granger test uses a parametric, augmented Dickey-Fuller (ADF) approach to accounting for serial correlation in the residual series.

The Engle-Granger test estimates a p-lag augmented regression of the form

(1)

Where u_{t} represents the residual of OLS regression, is the
difference of residual and p is then number of lagged differences chosen
to remove any evidence of serial correlation in the residuals. Two standard ADF test statistics consider, the one based on the τ-statistic
(tau) for testing the null hypothesis of nonstationary () and the other
based directly on the normalized autocorrelation coefficient ():

(3)

(4)

Where se() is the usual OLS estimator of the standard error of the
estimated The null hypothesis (H0) of no cointegration in equation
(4) is* ρ = 1 *and the alternative one (Ha) is* ρ < 1. *The lag length in the
model was determined using Akaike Information and Bayesian model
selection criteria, and then the estimate ADF statistic compared with
the critical values:

If ADF_{cal} > ADF_{critical} reject H_{0}: u_{t} is stationary, and then Yt_{1} and Yt_{2} are cointegrating.

If ADF_{cal} ≤ ADF_{critical} accept H_{0}: u_{t} is not stationary, and then Yt_{1} and Yt_{2} not cointegrated. If the hypothesis of no cointegration is
rejected i.e. long run relationship exists between the variable, an error
correction model (ECM) developed by Engle and Granger [12] would
be estimated, it considers bivariate market cointegration between any
pairs of markets i and j. The probability values were derived from the
Davidson and MacKinnon response surface simulation results

Error correction model (ECM) is a time series model in first
differences that contains an error correction term, which works to
bring two I(1) series back into long-run equilibrium (Wooldridge). To
learning about a potential long-run relationship between two series,
the concept of cointegration enriches the kinds of dynamic models at
our disposal. If y_{t} and x_{t} are I(1) processes and are not co integrated, a
dynamic model might be estimated in first differences as considered in
the following derivation.

Assuming the following two variables cointegration regression model:

Y_{t} = α + βX_{t} + u_{t} (5)

The Engle-Granger residual-based test for cointegration is simply unit root test applied to the residuals obtained from OLS estimation of the above Equation in two steps:

Step (1) regress Y on X in level to obtain the cointegration vector which is the predicted equilibrium relationships.

From step (1), u_{t} = (Y_{t} - α - βX_{t}) =the error term.

This is not for causal inference, but a necessary prerequisite. If and
only if u_{t} is stationary, we can proceed to step (2). If u_{t} is not stationary
then the Y,X relationship is spurious, not cointegrating.

Step (2) modifies the model in (5) to be:

Y_{t} = α_{0} + α_{1}Y_{t-1} + β_{0}X_{t} + β_{1}X_{t-1} + u_{t} (6)

Assume Y_{t} and X_{t} ∼I(1)

Subtract Y_{t-1} from both sides of equation and get:

ΔY_{t} = α_{0} + ρ_{1}Y_{t-1} + β_{0}X_{t }+ β_{1}X_{t-1} + u_{t}

Where ρ_{1} = (α_{1} -1)

Now add: β_{0}X_{t-1} - β_{0}X_{t-1} and get:

ΔY_{t} = α_{0 }+ ρ_{1}Y_{t-1} + β_{0}ΔX_{t }+ θ_{1}X_{t-1} - β_{0}X_{t-1} + u_{t} (7)

Where θ_{1} = (β_{1 }+ β_{0})

If ΔY_{t} stationary, Y_{t }and X_{t }cointegrated, then u_{t} must be I(0) as
well. Now from equation (5) the error term u_{t} = (Y_{t }- α - βX_{t}) then u_{t-1} =
(Y_{t-1} - α - βX_{t-1}) which is the error correction mechanism, then the error
correction model is:

ΔY_{t} = α_{0 }+ β_{0}ΔX_{t} + π(Y_{t-1} - α - βX_{t-1})+ e_{t}

Then ΔY_{t} = α_{0} + β_{0}ΔX_{t }+ π u_{t-1}+ e_{t } (8)

Equation (8) is the error correction model implies the last step of Engle and Granger cointegration test, and accordingly;

Coefficient on ΔX_{t} will tap short-run effect.

Negative coefficient on u_{t-1} will be error correction.

In equation (8) π<0; If y_{t-1} > βx_{t-1}, then y in the previous period
has overshot the equilibrium because π<0, the error correction term
works to push y back toward the equilibrium. Similarly, if y_{t-1}> βx_{t-1},
the error correction term induces a positive change in y back toward
the equilibrium.

In order to test stationarity of cattle prices for the markets
considered in this study, three approaches were applied to prices in
level. These are Dickey-Fuller and its augmentation (DF/ADF) test,
Phillips (PP) procedure and panel unit root test using the E-Views
software computer program [13]. Panel unit root tests provide an
overall aggregate statistic to examine whether there exists a unit root
in the pooled cross-section time series data and judge the time series
property of the data accordingly. Therefore, the panel-based unit root
tests have higher power than unit root tests based on individual time
series. The automatic selection methods choose to minimize one of
the following criteria: Akaike (AIC), Schwarz (SIC) and Hannan and
Quinn (HQ) selection criteria [14-16]. Here in this test three panel
unit root tests would be used, which were the Levin, Lin test, Fisher
ADF test and Fisher-PP test [17-19]. The results for these tests in levels
and first differences for selected markets during the periods (1995m1-
2011m12) are reported in **Table 1**.

Tests | Levels | First Differences | ||
---|---|---|---|---|

test statistic | the p-values | test statistic | the p-values | |

Levin et al. | -0.7785 | 0.2181 | -30.7911 | 0 |

ADF - Fisher | 6.32811 | 0.787 | 595.795 | 0 |

PP - Fisher | 5.62558 | 0.8457 | 796.4 | 0 |

Source: calculated from Appendix (A) using E-Views software computer programs.

**Table 1: **The Panel unit root test (in levels and First Differences) in selected markets (1995m1-2011m12).

Automatic lag length selection based on Schwarz information criteria (SIC).

It was clear that in case of the test in level all the results indicate the presence of unit root, as the LLC test and both Fisher tests fail to reject the null hypothesis of non-stationary (presence of unit root) in view of p-values which were non-significant. On the other hand in case of the first differences test the null hypothesis was rejected in all three tests according to the p-values which were significant at less than 1% level.

The results of other tests shows that all price series are nonstationary in level, while it was stationary in first differences for all variables, so all prices were integrated of order I(1).

Engle and Granger note that a linear combination of two or more I(1) series may be stationary, or I(0), in which case it might be said that the series are co integrated. Such a linear combination defines a cointegrating equation with cointegrating vector of weights characterizing the long-run relationship between the variables.

The Engle-Granger residual-based test for cointegration is simply unit root test applied to the residuals obtained from OLS estimation of Equation (1). Under the assumption that the series are not co integrated, all linear combinations including the residuals from OLS, are unit root nonstationary. Therefore, a test of the null hypothesis of no cointegration against the alternative of cointegration corresponds to a unit root test of the null of nonstationary against the alternative of stationarity. Given the five markets Elobied, Omdurman, Medani, Sennar and Nyala, twenty pairwise comparisons are possible for the two direction of dependency in case of cattle prices for the period 1995m1- 2011m12. For Engle-Granger test, two standard ADF test statistics consider, the one based on the τ-statistic (tau) for testing the null hypothesis of nonstationary and the other based directly on the Normalized Autocorrelation Coefficient (). The lag length in the model was determined using Akaike Information and Schwarz Bayesian model selection criteria, and then the estimated ADF statistic compared with the critical values.

The Engle-Granger method involves firstly running a cointegration
regression of one variable on another, and secondly checking whether
the regression residual from the first step is stationary using an ADF
test. **Table 2 **below reports twenty pairwise cointegration regression
results for cattle prices in the selected markets. Estimated parameters
(constant- explanatory) are presented with corresponding p-value
between two brackets for t-ratio. Goodness of fit (R2) and Cointegration
Durbin-Watson statistic (CRDW) also presented in **Table 2 **below.

Dependent variable | Independent variable | Parameter estimates | R^{2} |
CRDW | |
---|---|---|---|---|---|

Constant | Explanatory | ||||

Elobied | Omdurman | 1.17701 (0.0526) | 0.51356 | 0.11156 | 0.62771 |

-0.0033 | |||||

Elobied | Medani | 2.04265 (0.0000) | 0.28584 | 0.10716 | 0.57214 |

-0.0208 | |||||

Elobied | Sennar | 1.46704 (0.0024) | 0.47937 | 0.10962 | 0.58479 |

-0.0019 | |||||

Elobied | Nyala | 2.60466 | 0.130243 | 0.01201 | 0.52334 |

0 | -0.3968 | ||||

Omdurman | Elobied | 2.82516 (0.0000) | 0.221297 | 0.11281 | 0.71619 |

-0.0053 | |||||

Omdurman | Medani | 2.66042 | 0.254405 | 0.16913 | 0.71558 |

0 | -0.001 | ||||

Omdurman | Sennar | 2.67055 | 0.259463 | 0.09664 | 0.65681 |

0 | -0.0113 | ||||

Omdurman | Nyala | 2.60217 | 0.317079 | 0.17045 | 0.71482 |

0 | -0.0004 | ||||

Medani | Elobied | 2.23178 | 0.335492 | 0.08201 | 0.30873 |

0 | -0.01 | ||||

Medani | Omdurman | 0.87348 (0.1582) | 0.675964 | 0.142 | 0.37479 |

-0.0002 | |||||

Medani | Sennar | 0.79714 | 0.777497 | 0.28088 | 0.35992 |

-0.076 | 0 | ||||

Medani | Nyala | 1.88094 | 0.48569 | 0.10373 | 0.32449 |

0 | -0.0013 | ||||

Elobied | 2.33876 (0.0000) | 0.265052 | 0.11176 | 0.51471 | |

-0.0045 | |||||

Sennar | Omdurman | 1.93225 (0.0001) | 0.342617 | 0.09625 | 0.49159 |

-0.0125 | |||||

Sennar | Medani | 1.84181 (0.0000) | 0.398235 | 0.3034 | 0.54402 |

0 | |||||

Sennar | Nyala | 2.71452 | 0.148012 | 0.0321 | 0.46556 |

0 | -0.2093 | ||||

Nyala | Elobied | 2.46663 | 0.102208 | 0.01299 | 0.31175 |

0 | -0.3845 | ||||

Nyala | Omdurman | 0.5587 | 0.635222 | 0.15819 | 0.45271 |

-0.2889 | 0 | ||||

Nyala | Medani | 1.60984 | 0.359288 | 0.12776 | 0.38148 |

0 | -0.0011 | ||||

Nyala | Sennar | 2.06271 | 0.226334 | 0.03234 | 0.32143 |

0 | -0.1278 |

Source: Own author calculation.

**Table 2:** Cointegration regression results for cattle prices 1995M1- 2011M12.

The interesting in these results is that, in all cases R2<DW which means that, they do not suffer any spurious regression. According to Granger and Newbold (1974), an R2>DW is a good rule of thumb to suspect that the estimated regression is spurious. With regard to the goodness of fit (R2) for these relationships, it ranged from 1% low in case of Elobied on Medani and Nyala on Elobied to 30% high in case of Sennar on Medani relationship. That means the explanatory power to illustrate the strength of relationship is very weak. With regard to the variables coefficient there were sixteen out of twenty relationship were significant, these relations could be summarized in term of effectiveness, following, Omdurman and Medani markets affected by all others markets, while, Nyala market does not connected with Elobied and Sennar markets but it was connected to other markets. An important notice in all these relationship the directions was positive, that is to say an increase in prices in one market lead to an increase in the other market and the vice versa.

CRDW is the cointegration regression Durbin-Watson

- The figures in parentheses in column 3 and 4 stand for p-values of t-ratio.

According to Omdurman on Elobied and Omdurman on Nyala, the relationship between the two markets is significant at less than 1% level of significance, the direction between Omdurman and these two markets was positive which meant that an increasing in cattle prices of Elobied and Nyala markets leads to increase in cattle prices of Omdurman market which is logic result in sense that Elobied and Nyala markets are considered as supply markets while Omdurman is demand one. Moreover, the goodness of fit show that about 11% changing in Omdurman market prices causes by the changing in Elobied market prices, whereas 89% of the changing causes by changing in other markets.

The Engle-Granger residual-based test for cointegration is simply
unit root test applied to the residuals obtained from OLS estimation
and the results reported in **Table 3**.

Models | Engle-Granger tau-statistic | Engle-Granger z-statistic | Residuals Stationarity Test | |
---|---|---|---|---|

1 | Elobied on Omdurman | -6.08249 (0.0000) | -63.3895 (0.0000) | -0.31226 (0.0000) |

2 | Elobied on Medani | -5.80071 (0.0000) | -58.53103 (0.0000) | -0.28833 (0.0000) |

3 | Elobied on Sennar | -5.84350 (0.0000) | -59.10871 (0.0000) | -0.29117 (0.0000) |

4 | Elobied on Nyala | -5.49880 (0.0000) | -53.29833 (0.0000) | -0.26255 (0.0000) |

5 | Omdurman on Elobied | 6.67538 (0.0000) | -73.37765 (0.0000) | -0.36146 (0.0000) |

6 | Omdurman on Medani | -6.70701 (0.0000) | -73.48324 (0.0000) | 0.36198 (0.0000) |

7 | Omdurman on Sennar | -6.30781 (0.0000) | -66.81507 (0.0000) | -0.32913 (0.0000) |

8 | Omdurman on Nyala | -6.64597 (0.0000) | -72.87077 (0.0000) | -0.35896 (0.0000) |

9 | Medani on Elobied | -4.54545 (0.0014) | -34.38203 (0.0018) | -0.16937 (0.0000) |

10 | Medani on Omdurman | -4.89959 (0.0004) | -39.29877 (0.0005) | -0.19359 (0.0000) |

11 | Medani on Sennar | -4.86573 (0.0004) | -38.52192 (0.0006) | -0.18976 (0.0006) |

12 | Medani on Nyala | -4.64131 (0.0000) | -35.51242 (0.0010) | -0.17493 (0.0014) |

13 | Sennar on Elobied | -5.46043 (0.0000) | --5.460431 (0.0000) | -0.25757 (0.0000) |

14 | Sennar on Omdurman | -5.31116 (0.0001) | -49.95801 (0.0000) | -0.24609 (0.0000) |

15 | Sennar on Medani | -5.70327 (0.0000) | -55.85167 (0.0000) | -0.27513 (0.0000) |

16 | Sennar on Nyala | -5.15169 (0.0001) | -47.25617 (0.0001) | 0.23278 (0.0000) |

17 | Nyala on Elobied | -3.22374 (0.0703) | -21.69171 (0.0346) | --0.12619 (0.0015) |

18 | Nyala on Omdurman | -3.46228 (0.0393) | -25.15623 (0.0159) | -0.15321 (0.0007) |

19 | Nyala on Medani | -3.55693 (0.0307) | -26.36532 (0.0121) | 0.15256 (0.0005) |

20 | Nyala on Sennar | -3.19928 (0.0744) | -21.44032 (0.0365) | --0.12645 (0.0016) |

Source: Own author calculation.

**Table 3: **Engle-Granger cointegration results for cattle prices in selected markets, 1995M1- 2011M12

The tests statistics values, residuals stationarity test and their
corresponding p-values under 5% significance levels, are reported
in the **Table 3 **above. The lag length in the model was determined
automatically using Akaike Information and Schwarz Bayesian model
selection criteria which is available in E-Views software program
automatically.

As to the tests themselves, the Engle-Granger tau-statistic
(t-statistic for ADF test equation for residuals) and normalized
autocorrelation coefficient (which was termed as z-statistic) both reject
the null hypothesis of no cointegration (unit root in the residuals) at
the 1% level in all cases. On balance, the evidence clearly suggests that
these markets were co integrated. Once the cointegration properties
were found in price series, the second step to the Engle-Granger test is
the residuals stationarity test. It is found obvious from **Table 3 **that all
residuals stationarity tests were rejected.

- The figures in parentheses of **Table 3 **above are p-values.

The nonstationary at the 1% level and accept that all the residuals from OLS regression were stationary and therefore, the above relations were long run, and then the error correction model (ECM) would be estimated to integrate the dynamics of short run with long run adjustment process.

Following Babiker [19] once cointegration relationship is established, the validity of an error correction model, using residuals from these cointegration regressions, was to imply that the variables in concern were co integrated as suggested by Engle and Granger representation theorem. The model may be interpreted as possessing long run equilibrium, although random shocks push the system away from equilibrium. In the short run, the error correction term, therefore, causes changes in the variables of the model. In the error correction model the dynamics of the both short run and long run adjustment processes were modeled simultaneously.

As a pre-conditions for estimation of the model describing the error correction model of equation (8), that the cointegration relationship
should be establish between the variables and the error term ut is
stationary. As it’s presented in results above all these two conditions
were satisfied, then the model in equation (8) was estimated using OLS
technique. The results are summarised in **Table 4**.

- The figures in parentheses are p-values.

As shown in **Table 4 **above, analysis of the cattle prices would
suggest that the error correction model test should be conducted
under the assumption of having linear data trend in the series and thus
allowing constant and trend (equation (8) in the methodology). The
error correction coefficients in the ECM equations were significant at
1% level and associated with the desirable negative signs in all cases,
this shows that the error correction mechanism adjusts significantly
to shocks to its equilibrium relationship with its hypothesized determinants that are caused by exogenous changes in past values.
The goodness of fit fluctuated between 5% in case of Nyala on Sennar
and 27% in case of Sennar on Medani. While Durbin-Watson statistic
explains the absence of autocorrelation problem as its values close to 2.

Models | EstimatedÂ Coefficient | R-squared | Durbin-Watson stat | ARCH test | |
---|---|---|---|---|---|

D(independent variable) | RESID(-1) | ||||

Elobied on Omdurman | 0.180198 | -0.135565 (0.0026) | 0.100681 | 2.266805 | 39.50262 (0.0000) |

-0.0184 | |||||

Elobied on Medani | 0.234308 | -0.127992 (0.0032) | 0.1122 | 2.309869 | 38.69968 (0.0000) |

-0.0032 | |||||

Elobied on Sennar | 0.255551 | -0.143581 | 0.127078 | 2.232592 | 38.89655 (0.0000) |

-0.0006 | (0. 0007) | ||||

Elobied on Nyala | 0.20907 | -0.124692 (0.0016) | 0.097452 | 2.195357 | 28.22790 (0.0000) |

-0.0118 | |||||

Omdurman on Elobied | 0.111938 | -0.224177 (0.0000) | 0.162919 | 2.206571 | 0.686804 (0.4082) |

-0.0054 | |||||

Omdurman on Medani | 0.254339 | -0.227075 (0.0000) | 0.194984 | 2.243806 | 0.038302 |

0 | -0.845 | ||||

Omdurman on Sennar | 0.229828 | -0.208174 (0.0000) | 0.185247 | 2.211422 | 0.692262 |

0 | -0.4064 | ||||

Omdurman on Nyala | 0.292985 | -0.242575 (0.0000) | 0.202505 | 2.186225 | 0.945908 |

0 | -0.3319 | ||||

Medani on Elobied | 0.154465 | -0.058734 (0.0437) | 0.078648 | 2.064032 | 0.170445 |

0 | -0.6802 | ||||

Medani on Omdurman | 0.227881 | -0.067527 (0.0309) | 0.104036 | 2.035329 | 0.006442 |

0 | -0.9361 | ||||

Medani on Sennar | 0.417013 | -0.081341 (0.0133) | 0.244517 | 2.121785 | 0.067652 |

0 | -0.7951 | ||||

Medani on Nyala | 0.165365 | -0.074024 (0.0084) | 0.074997 | 1.935802 | 0.12776 |

-0.0025 | -0.7211 | ||||

Sennar on Elobied | 0.170585 | -0.136114 (0.0001) | 0.129295 | 0.129295 | 10.83712 |

0 | -0.0199 | ||||

Sennar on Omdurman | 0.23145 | -0.127665 (0.0001) | 0.138253 | 2.138849 | 23.9457 |

0 | -0.0008 | ||||

Sennar on Medani | 0.47167 | -0.151289 (0.0001) | 0.270715 | 2.264598 | 47.29401 |

0 | 0 | ||||

Sennar on Nyala | 0.157787 | -0.129306 (0.0000) | 0.112742 | 2.085724 | 7.83951 |

-0.0024 | -0.014 | ||||

Nyala on Elobied | 0.096076 | -0.043953 (0.1216) | 0.055564 | 2.464118 | 0.058792 |

-0.0035 | -0.8396 | ||||

Nyala on Omdurman | 0.214323 | -0.065698 (0.0437) | 0.103168 | 2.537321 | 33.9508 |

0 | -0.0069 | ||||

Nyala on Medani | 0.119755 | -0.057936 (0.0714) | 0.061935 | 2.481989 | 0.727948 |

-0.0141 | -0.3946 | ||||

Nyala on Sennar | 0.145063 | -0.045169 (0.1170) | 0.057617 | 2.492298 | 0.269864 |

-0.0012 | -0.604 |

Source: Own author calculation.

**Table 4: **Estimated Error Correction Model for Cattle Prices, 1995M1-2011M2.

A significant coefficient in seventeen relations of the error correction terms provides evidence that the two markets prices were co integrated and that they share a long-run common trend. Examination of these coefficients in each equation provides information as to the degree and direction of the adjustment back towards the long-run equilibrium. There were three markets relation found to be non-significance which were Nyala on Elobied and Nyala on Sennar.

Hence a positive departure from equilibrium of the two markets
prices in the previous period will be corrected by a negative coefficient
of error correction term. Thus, the long-run relationship between the
two markets prices is stable over time. The stable markets prices provide
important information for long-run strategic planning. In **Table 4 **above, the estimated coefficients for ECM is fluctuated between 4% in
case of Nyala on Elobied and 24% in case of Omdurman on Nyala, and
it was significant at 1% level, suggesting that the last period (month)
disequilibrium in prices of Nyala on Elobied, for example, corrected
in the next month by 4%, where it seems to adjust slowly towards the
long-run equilibrium, On the other hand, with respect to Omdurman
on Nyala prices relationship, the last period (month) disequilibrium
in prices corrected in the next month by 24%. However both values
seem to adjust slowly towards the long-run equilibrium. This implies
that any shock that forces prices from their long-run value would take
a long time for prices to return to its equilibrium unless there are other
shocks that counter the initial one. The short-run effect presented by
the coefficient of independent variables in these models, which had
positive signs and significant at all cases. This finding suggests that the
short-run changes in independent market have a positive impact on
short-run changes in dependent market with the value of coefficient.
i.e. in the short-run Nyala cattle market affected Omdurman cattle
market by 29% and so on with respect to others models.

Considering the model without constant and trend, summing up
the finding in the three unit root tests applied in this study, it’s clear
that all prices series were non-stationary in level, while they were
stationary in first differences for all variables. From these results it
could be concluded that all prices were integrated of order I (1). As
the long run analysis of cattle prices in selected markets indicated, a
strong evidence of cointegration of pairs of markets exists as shown in
**Figure 1**. The error correction coefficients in the ECM equations show
that the error correction mechanism adjusts significantly to shocks to
its equilibrium relationship with its hypothesized determinants that are
caused by exogenous changes in past values. The estimated coefficients
for ECM were fluctuated between 4% and 24% and significant at 1%
level, suggesting slow adjustment towards the long-run equilibrium.
This implies that any shock that forces prices from their long-run value
will take a long time for prices to return to its equilibrium, that means
the speed of adjustment was highest in case of Omdurman on Nyala
and lowest in case of Nyala on Elobied, this might be due to supply and
demand relation between Nyala and Omdurman in sense that Nyala
and Elobied were supply markets.

**Figure 1 **reveals that the prices of cattle markets had effect on each
other’s in the short run, i.e. any market was affected by other markets
and has effect on other markets. This result seems to be reasonable,
because of the recent paved roads that linked between all these markets which accelerated and facilitated the movement between these markets,
addition to a huge capital that earmarked to livestock **business **recently.
Babiker [19] found that Omdurman prices were affected by its own
prices and exert short effects on all other markets while Medani and
Elobied exert short run effects on Omdurman but not Nyala.

- Goodwin BK, Schroeder TC (1991) Cointegration tests and spatial price linkages in regional cattle markets. American Journal of Agricultural Economics 73: 452-464.
- Barrett CB, Jau RL, DeeVon B (2000) Factor and product market tradability and equilibrium in Pacific Rim pork industries. Journal of Agricultural and Resource Economics 25: 68-87.
- Babiker BI, Abdalla AGM (2009) Spatial price transmission: A study of sheep markets in Sudan.African Journal of Agricultural and Resource Economics 3: 43-56.
- Johansen S (1988) Statistical analysis of cointegration vectors. Journal of economic dynamics and control 12: 231-254.
- Ibrahim AA (1999) The Development of the Livestock Sector in Sudan: A Case Study of Public Policy Analysis. OSSEREAâ€™s research report series, Addis Ababa.
- Idris B (1986) Marketing System. In Zahlan A B (edn): Agricultural Sector of the Sudan: Policy and Systems Studies, London: 358-379.
- M.SA (1986) Livestock Farming Systems In: Zahlan, AB (edn): Agricultural Sector of the Sudan: Policy and Systems Studies, London 215-238.
- El Agip FM (2001) Marketing of Livestock in the Sudan: An Analysis of its Efficiency. Unpublished Ph.D. Thesis: University of Khartoum, Sudan.
- Babiker NM (2006) Livestock Markets in the Sudan: A cointegration Approach, Ph D Thesis, Gazira University.
- Dickey DA, Fuller WA (1981) Likelihood ratio statistics for autoregressive time series with a unit root. Econometrica: Journal of the Econometric Society 49: 1057-1072.
- Phillips PC, Perron P (1988) Testing for a unit root in time series regression. Biometrika 75: 335-346.
- Engle RF, Granger CWJ (1987) Co-integration and error correction: representation, estimation, and testing. Econometrica: Journal of the Econometric Society 55: 251-276.
- McKenzie CR, Takaoka S (2012) EViews 7.2. Journal of applied econometrics 27: 1205-1210.
- Akaike H (1987) Factor analysis and AIC. Psychometrika 52: 317-332.
- Schwarz G (1978) Estimating the dimension of a model. The annals of statistics 6: 461-464.
- Hannan EJ, Quinn BG (1979) The determination of the order of an autoregression. Journal of the Royal Statistical Society. Series B (Methodological) 41: 190-195.
- Levin A, Lin, CF, James C, Chia S (2002) Unit root tests in panel data: asymptotic and finite-sample properties. Journal of econometrics 108: 1-24.
- Fisher RA (1932) Statistical Methods for Research Workers, (5th edn), Edinburgh, Oliver and Boyd, UK.
- Babiker NMM, BusharaMOA (2006) Integration Between Cattle Markets in the Sudan, 1980-2000. Gezira Journal of Agricultural Science 4: 130-146.

Select your language of interest to view the total content in your interested language

- Accountancy and Finance
- Accounting Information
- Accounting Review
- Applied Economics
- Assessment Scales
- Avenues of Investment
- BCM
- Banking Research
- Banking Research Studies
- Business
- Business Cycle
- Business Management
- Business Plan
- Business Research
- Business Theory
- CRM
- Capital Markets
- Capital Movements
- Capital Structure
- Chief Marketing Officer
- Classical Economics
- Computable General Equilibrium Model
- Corporate Finance
- Corporate Governance Structure
- Corporate governance system
- Demand Theory
- Development Economics
- E-Governance
- E-Retailing Market
- E-banking
- Econometrics
- Economic Capital
- Economic Cycle
- Economic Growth
- Economic Policies
- Economic Resources
- Economic Transparency
- Economic indicator
- Economics Studies
- Electronic Business
- Electronic Commerce
- Empirical Analysis
- Entrepreneurial Development
- Entrepreneurship
- Finance and accounting
- Financial Affairs
- Financial Crisis
- Financial Economics
- Financial Management services
- Financial Markets
- Financial Reporting Standard
- Financial Risk
- Financial Services
- Financial accounting
- Financial and Nonfinancial Information
- Financial plan
- Fiscal and tax policies
- Foreign Exchange
- Game theory
- General finance
- Global Accounting
- Global Market
- Gross Domestic Product -GDP
- HRM
- Health Management
- Hospitality Management
- Human Capital
- Income Smoothing
- Indexation
- Industrial Business
- Industrial and Management Optimization
- Innovation Policy and the Economy
- Intellectual Capital Disclosures
- International Business
- Logistics management
- Management
- Management Accounting
- Management in Education
- Managerial Finance
- Managerial accounting
- Marginal Utility
- Market Analysis
- Market Equilibrium
- Marketing Analysis
- Marketing management
- Marketing-Accounting-Finance Interface
- Micro Economics
- Microfinance
- Monetary Neutrality
- Monetary Policy
- Multinational finance
- Nasdaq
- New Economy
- Organizational studies
- Panel Data
- Parameter Estimation
- Primary Market
- Profitability
- Risk management
- Secondary Market
- Small Firms
- Small Scale Business
- Social Economics
- Socio-Economic Planning Sciences
- SocioEconomics Status
- Stock Market
- Stock Market Returns
- Stock Return Predictability
- Strategic management
- Time Series
- Total Quality Management (TQM)
- Trading
- Value based Management
- Wealth Management
- Welfare Economics
- World banking

- Total views:
**8044** - [From(publication date):

April-2016 - Oct 23, 2017] - Breakdown by view type
- HTML page views :
**7890** - PDF downloads :
**154**

Peer Reviewed Journals

International Conferences 2017-18