Academia.eduAcademia.edu

Figure 2 – uploaded by Richard K Bannor

See full PDFdownloadDownload figure
A test for a suitable lag length to be included in the co-integration test was performed, because the results of co- integration tests can be quite sensitive to this Hafer and Sheehan, 1991 and Hai etal., 2004). The number of lags was selected by applying five different multivariate lag selection criteria: the Akaike information criterion (AIC), the Hannan-Quin information criterion (HQIC), and the Schwarz’s Bayesian information criterion (SBIC), FPE and LR. A vector autoregression (VA R) on the differenced series was conducted and lag length of the model with the least AIC, HQIC and SBIC values chosen as the appropriate lag length to be included in the cointegration test. The results of co-integration test as given in the Table 1 indicate the right lag length for the analysis was 5 lags.

Table 1 A test for a suitable lag length to be included in the co-integration test was performed, because the results of co- integration tests can be quite sensitive to this Hafer and Sheehan, 1991 and Hai etal., 2004). The number of lags was selected by applying five different multivariate lag selection criteria: the Akaike information criterion (AIC), the Hannan-Quin information criterion (HQIC), and the Schwarz’s Bayesian information criterion (SBIC), FPE and LR. A vector autoregression (VA R) on the differenced series was conducted and lag length of the model with the least AIC, HQIC and SBIC values chosen as the appropriate lag length to be included in the cointegration test. The results of co-integration test as given in the Table 1 indicate the right lag length for the analysis was 5 lags.

Related Figures (6)

The study first examined each price series for evidence of non-stationarity in order to confirm that co-integration approach. At level 0, Kasoa, Swedru and Bawjiase markets cassava wholesale prices were not stationary but Ajumako and Cape Coast prices were stationary. DF-GLS test, Augmented Dickey Fuller (A DF) and Philips-Perron (PP) showed similar results as indicated in Table 2.
The study first examined each price series for evidence of non-stationarity in order to confirm that co-integration approach. At level 0, Kasoa, Swedru and Bawjiase markets cassava wholesale prices were not stationary but Ajumako and Cape Coast prices were stationary. DF-GLS test, Augmented Dickey Fuller (A DF) and Philips-Perron (PP) showed similar results as indicated in Table 2.
source: Author s Computation At rank 0: Ho: There is no co-integration between the various market prices H1: There is co-integration between the various market prices. NB: We accept null hypothesis when trace statistics or max statistics is less than the 5% Critical value at rank 0. At rank 1,3: Ho: There is (1) or (3) co-integration of the various markets at rank 1 or 3 H1: There is no 1 or 3 co-integration of the markets at rank 1 or 3. NB: We accept null hypothesis when trace statistics or max statistics is less than the 5% Critical value at rank 1 or 3
source: Author s Computation At rank 0: Ho: There is no co-integration between the various market prices H1: There is co-integration between the various market prices. NB: We accept null hypothesis when trace statistics or max statistics is less than the 5% Critical value at rank 0. At rank 1,3: Ho: There is (1) or (3) co-integration of the various markets at rank 1 or 3 H1: There is no 1 or 3 co-integration of the markets at rank 1 or 3. NB: We accept null hypothesis when trace statistics or max statistics is less than the 5% Critical value at rank 1 or 3
The co-integration tests results as shown in the Table 4 indicates three co-integration between two markets pairs such as Ajumako- Kasoa, Swedru-Cape Coast and Ajumak- Swedru at Rank Order 1. This means that, the prices of the two markets move closely together in the long run although in the short run they
The co-integration tests results as shown in the Table 4 indicates three co-integration between two markets pairs such as Ajumako- Kasoa, Swedru-Cape Coast and Ajumak- Swedru at Rank Order 1. This means that, the prices of the two markets move closely together in the long run although in the short run they
ource: Author's computation |: No short run causality running from market A to B [: Short run causality running from A to B or market A causes price changes in market 1B: Reject null hypothesis when the p value is > 5%
ource: Author's computation |: No short run causality running from market A to B [: Short run causality running from A to B or market A causes price changes in market 1B: Reject null hypothesis when the p value is > 5%
JUULILE, AULNUL 5 UUIIVPULALLUEE Langrange Multiplier Test; Ho: No autocorrelation; H1: There is autocorrelation Jarque Bera Test: Ho: Residuals are normally distributed; H1: Residuals are not normally distributed NB: Reject null hypothesis when p value is less than 5% Table 7: Diagnostic checking of the VAR and VEC models
JUULILE, AULNUL 5 UUIIVPULALLUEE Langrange Multiplier Test; Ho: No autocorrelation; H1: There is autocorrelation Jarque Bera Test: Ho: Residuals are normally distributed; H1: Residuals are not normally distributed NB: Reject null hypothesis when p value is less than 5% Table 7: Diagnostic checking of the VAR and VEC models
Related topics:
Spatial price transmission
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved