Rice Genomics and Genetics 2015, Vol.6, No.5, 1-10
6
Table 3 Results of Econometric Integration of Price Series
Variables
Price level 1(0)
First difference 1(1)
Market price series
ADF statistics
PP statistics
ADF statistics
PP statistics
Ekiti
-1.4409
-1.6042
-11.5337
-19.2139
Lagos
-1.0567
-1.7997
-8.8527
-41.2822
Ogun
-1.3519
-1.5975
-18.9991
-30.5026
Ondo
-1.5214
-1.6406
-11.6777
-18.4598
Osun
-1.2456
-2.2268
-8.9768
-43.6053
Oyo
-0.9827
-0.9979
-14.7398
-14.7208
Notes: Critical values are -3.4870 and -3.4861 at the 99% and -2.8859 and -2.8861 at the 95%. Confidence levels for price level and
first difference series, respectively. Source: Compiled from results of stationarity test. If the absolute value of the ADF or PP is lower
than their critical statistics, we fail to reject the null hypothesis of non-stationarity. Confidence levels for price level and first
difference series, respectively. Source: Compiled from results of stationarity test. If the absolute value of the ADF or PP is lower than
their critical statistics, we fail to reject the null hypothesis of non-stationarity.
As discussed in the methodology section, this meant
that all price data were integrated of order one I(1), a
requirement for Johansen’s co-integration analysis
(Johansen and Juselius, 1990; Juselius, 2006).
To bolster our findings concerning the I(1) and I(0)
nature of the price series at their levels and their first
differences, respectively, the Phillip-Perron (PP) test
was also conducted. The PP test, like the ADF test,
indicated significance for all variables, rejecting the
null hypothesis of stationarity at the 1% and 5% levels
of significance. The findings here were in accordance
with earlier findings and conclusion that food
commodity price series are mostly stationary of order
one i.e. I(1) (Mafimisebi, 2001; Okoh and Egbon,
2003; Mafimisebi 2008, 2012). According to
Mafimisebi (2012), the result is probably explained by
the fact that most food price series have trends in them
owing to inflation which culminates in such data
exhibiting mean non-stationarity.
4.1.5. Results of Long-run Integration Test
The co-integration test results for market pairs were
presented in Table 4. Generally, the results indicated
price co-integration at 5% level of significance by
rejecting the null hypothesis in favour of the alter-
native, in fourteen (14) out of fifteen (15) market pairs.
Table 4 Result of Pair-wise Co-integration Test of Imported Rice Market
Market Pairs
Trace Test Statistics
Maximal Eigenvalue Test Statistics
Pi-Pj
Lagos/Ogun
23.202**
21.580**
Lagos/Ondo
32.814**
30,488**
Lagos/Osun
32.670**
30.495**
Lagos/Oyo
44.403**
43.417**
Lagos/Ekiti
36.982**
34.581**
Ogun/Osun
17.928*
16.235*
Ogun/Oyo
15.734*
14.762*
Ogun/Ekiti
19.665*
17.747*
Ondo/Osun
42.905**
40.593**
Ondo/Oyo
25.799**
24.519**
Ondo/Ekiti
52.470**
50.049**
Osun/Oyo
27.985**
26.859**
Osun/Ekiti
Ekiti/Oyo
49.625**
34.806**
47.554**
33.484**
