11:12 p.m.
when i estimate the garch parameter of cj stock price during 1 year data
using gretl and R.
IN Gnu-R.
garch(x,c(1,1))
***** ESTIMATION WITH ANALYTICAL GRADIENT *****
I INITIAL X(I) D(I)
1 0.389259E-03 0.100E+01
2 0.500000E-01 0.100E+01
3 0.500000E-01 0.100E+01
IT NF F RELDF PRELDF RELDX STPPAR D*STEP
NPRELDF
0 1 -0.830E+03
1 8 -0.830E+03 0.43E-06 0.95E-06 0.1E-04 0.8E+09 0.1E-05
0.36E+03
2 18 -0.831E+03 0.13E-02 0.22E-02 0.5E+00 0.2E+01 0.1E+00
0.20E+00
3 21 -0.835E+03 0.48E-02 0.43E-02 0.7E+00 0.1E+01 0.4E+00
0.63E-01
4 23 -0.836E+03 0.19E-02 0.13E-02 0.4E-01 0.2E+01 0.4E-01
0.80E-01
5 26 -0.839E+03 0.27E-02 0.31E-02 0.9E-01 0.2E+01 0.1E+00
0.13E+01
6 27 -0.839E+03 0.88E-03 0.26E-02 0.8E-01 0.2E+01 0.1E+00
0.31E+00
7 36 -0.840E+03 0.50E-03 0.12E-02 0.6E-05 0.4E+01 0.8E-05
0.74E-02
8 37 -0.840E+03 0.29E-04 0.28E-04 0.5E-05 0.2E+01 0.8E-05
0.37E-02
9 38 -0.840E+03 0.36E-06 0.18E-05 0.5E-05 0.2E+01 0.8E-05
0.35E-02
10 46 -0.841E+03 0.11E-02 0.19E-02 0.5E-01 0.8E+00 0.7E-01
0.35E-02
11 48 -0.843E+03 0.22E-02 0.22E-02 0.4E-01 0.9E+00 0.7E-01
0.15E-01
12 50 -0.846E+03 0.41E-02 0.46E-02 0.7E-01 0.1E+01 0.1E+00
0.73E-01
13 52 -0.847E+03 0.14E-02 0.19E-02 0.2E-01 0.2E+01 0.5E-01
0.11E-01
14 53 -0.848E+03 0.14E-02 0.16E-02 0.2E-01 0.2E+01 0.5E-01
0.10E-01
15 55 -0.849E+03 0.23E-03 0.29E-03 0.3E-02 0.2E+01 0.5E-02
0.57E-02
16 58 -0.849E+03 0.90E-03 0.15E-02 0.2E-01 0.1E+01 0.3E-01
0.67E-02
17 60 -0.849E+03 0.55E-04 0.13E-03 0.4E-02 0.1E+01 0.1E-01
0.27E-03
18 62 -0.849E+03 0.24E-05 0.57E-05 0.8E-03 0.8E+00 0.2E-02
0.91E-05
19 64 -0.849E+03 0.11E-06 0.20E-05 0.3E-03 0.1E+01 0.7E-03
0.30E-05
20 65 -0.849E+03 0.23E-06 0.38E-06 0.1E-03 0.9E+00 0.3E-03
0.69E-06
21 78 -0.849E+03 -0.20E-12 0.48E-13 0.1E-13 0.2E+06 0.2E-13
0.23E-06
***** FALSE CONVERGENCE *****
FUNCTION -0.849489E+03 RELDX 0.114E-13
FUNC. EVALS 78 GRAD. EVALS 21
PRELDF 0.477E-13 NPRELDF 0.231E-06
I FINAL X(I) D(I) G(I)
1 0.167570E-05 0.100E+01 0.188E+04
2 0.630399E-01 0.100E+01 0.113E+01
3 0.944707E+00 0.100E+01 0.364E+00
Call:
garch(x = x, order = c(1, 1))
Coefficient(s):
a0 a1 b1
1.676e-06 6.304e-02 9.447e-01
Warning message:
NANs... in: sqrt(pred$e)
but in gretl
The convergence criterion was not met.
i read previous article 'convergence is not met' and i know two assistence.
first. Includeing suitable indepedent variable...
but i just want analysis only 245 data
second . transfomation
this data already log-diffrential transformation. how can i saw a result.
not adjust includeing variable.
thank u for reading my dumb mail. i wait you warth answer... ^^;
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE gretldata SYSTEM "gretldata.dtd">
<gretldata name="CJ" frequency="1" startobs="1"
endobs="247"
type="cross-section">
<variables count="3">
<variable name="price"
label="Stock Price"
displayname="Stock Price"
/>
<variable name="Return"
label="Return"
displayname="Return"
/>
<variable name="x"
label="x=obs"
/>
</variables>
<observations count="247" labels="false">
<obs>68300 -0.022207 1 </obs>
<obs>66800 0.008942 2 </obs>
<obs>67400 -0.007446 3 </obs>
<obs>66900 0.007446 4 </obs>
<obs>67400 0.004441 5 </obs>
<obs>67700 -0.010394 6 </obs>
<obs>67000 0.000000 7 </obs>
<obs>67000 -0.002990 8 </obs>
<obs>66800 -0.016604 9 </obs>
<obs>65700 -0.001523 10 </obs>
<obs>65600 0.000000 11 </obs>
<obs>65600 -0.004584 12 </obs>
<obs>65300 0.018210 13 </obs>
<obs>66500 0.004501 14 </obs>
<obs>66800 -0.012048 15 </obs>
<obs>66000 -0.007605 16 </obs>
<obs>65500 0.000000 17 </obs>
<obs>65500 -0.004591 18 </obs>
<obs>65200 0.000000 19 </obs>
<obs>65200 -0.003072 20 </obs>
<obs>65000 0.012233 21 </obs>
<obs>65800 0.013585 22 </obs>
<obs>66700 -0.016629 23 </obs>
<obs>65600 -0.009188 24 </obs>
<obs>65000 0.007663 25 </obs>
<obs>65500 0.000000 26 </obs>
<obs>65500 0.046242 27 </obs>
<obs>68600 0.015908 28 </obs>
<obs>69700 0.017070 29 </obs>
<obs>70900 0.000000 30 </obs>
<obs>70900 0.000000 31 </obs>
<obs>70900 0.008427 32 </obs>
<obs>71500 0.016644 33 </obs>
<obs>72700 0.025799 34 </obs>
<obs>74600 0.006680 35 </obs>
<obs>75100 -0.002667 36 </obs>
<obs>74900 0.043116 37 </obs>
<obs>78200 -0.012870 38 </obs>
<obs>77200 0.010309 39 </obs>
<obs>78000 -0.011606 40 </obs>
<obs>77100 -0.036992 41 </obs>
<obs>74300 0.026562 42 </obs>
<obs>76300 -0.019855 43 </obs>
<obs>74800 -0.004019 44 </obs>
<obs>74500 -0.006734 45 </obs>
<obs>74000 -0.004062 46 </obs>
<obs>73700 -0.033106 47 </obs>
<obs>71300 -0.015548 48 </obs>
<obs>70200 0.011332 49 </obs>
<obs>71000 0.025036 50 </obs>
<obs>72800 -0.004129 51 </obs>
<obs>72500 -0.020907 52 </obs>
<obs>71000 0.016760 53 </obs>
<obs>72200 -0.023827 54 </obs>
<obs>70500 -0.011412 55 </obs>
<obs>69700 -0.012997 56 </obs>
<obs>68800 -0.019076 57 </obs>
<obs>67500 0.021979 58 </obs>
<obs>69000 0.001448 59 </obs>
<obs>69100 0.031342 60 </obs>
<obs>71300 -0.009866 61 </obs>
<obs>70600 -0.015703 62 </obs>
<obs>69500 -0.007220 63 </obs>
<obs>69000 -0.010197 64 </obs>
<obs>68300 0.010197 65 </obs>
<obs>69000 0.029981 66 </obs>
<obs>71100 0.016737 67 </obs>
<obs>72300 0.000000 68 </obs>
<obs>72300 -0.002770 69 </obs>
<obs>72100 -0.006959 70 </obs>
<obs>71600 0.000000 71 </obs>
<obs>71600 -0.011236 72 </obs>
<obs>70800 0.000000 73 </obs>
<obs>70800 -0.011364 74 </obs>
<obs>70000 -0.037850 75 </obs>
<obs>67400 0.024910 76 </obs>
<obs>69100 -0.008721 77 </obs>
<obs>68500 -0.017674 78 </obs>
<obs>67300 -0.019505 79 </obs>
<obs>66000 0.029853 80 </obs>
<obs>68000 0.037522 81 </obs>
<obs>70600 0.000000 82 </obs>
<obs>70600 0.007057 83 </obs>
<obs>71100 0.016737 84 </obs>
<obs>72300 0.008265 85 </obs>
<obs>72900 0.006835 86 </obs>
<obs>73400 0.000000 87 </obs>
<obs>73400 -0.012337 88 </obs>
<obs>72500 0.000000 89 </obs>
<obs>72500 -0.008310 90 </obs>
<obs>71900 0.011065 91 </obs>
<obs>72700 0.004118 92 </obs>
<obs>73000 -0.002743 93 </obs>
<obs>72800 -0.011050 94 </obs>
<obs>72000 0.009675 95 </obs>
<obs>72700 0.000000 96 </obs>
<obs>72700 0.015017 97 </obs>
<obs>73800 -0.031661 98 </obs>
<obs>71500 0.019391 99 </obs>
<obs>72900 0.014976 100 </obs>
<obs>74000 -0.012237 101 </obs>
<obs>73100 0.029652 102 </obs>
<obs>75300 -0.003992 103 </obs>
<obs>75000 0.017185 104 </obs>
<obs>76300 -0.002625 105 </obs>
<obs>76100 -0.006592 106 </obs>
<obs>75600 -0.026811 107 </obs>
<obs>73600 0.000000 108 </obs>
<obs>73600 -0.008186 109 </obs>
<obs>73000 0.020340 110 </obs>
<obs>74500 0.002681 111 </obs>
<obs>74700 -0.009415 112 </obs>
<obs>74000 0.026668 113 </obs>
<obs>76000 0.014370 114 </obs>
<obs>77100 -0.001298 115 </obs>
<obs>77000 0.002594 116 </obs>
<obs>77200 -0.002594 117 </obs>
<obs>77000 -0.027652 118 </obs>
<obs>74900 -0.002674 119 </obs>
<obs>74700 -0.010767 120 </obs>
<obs>73900 -0.016371 121 </obs>
<obs>72700 0.004118 122 </obs>
<obs>73000 -0.002743 123 </obs>
<obs>72800 0.014997 124 </obs>
<obs>73900 -0.014997 125 </obs>
<obs>72800 -0.011050 126 </obs>
<obs>72000 -0.009770 127 </obs>
<obs>71300 0.005594 128 </obs>
<obs>71700 0.017969 129 </obs>
<obs>73000 -0.001371 130 </obs>
<obs>72900 0.008197 131 </obs>
<obs>73500 0.022866 132 </obs>
<obs>75200 -0.006671 133 </obs>
<obs>74700 -0.009415 134 </obs>
<obs>74000 -0.013606 135 </obs>
<obs>73000 -0.016575 136 </obs>
<obs>71800 0.000000 137 </obs>
<obs>71800 -0.006988 138 </obs>
<obs>71300 0.008380 139 </obs>
<obs>71900 -0.015417 140 </obs>
<obs>70800 -0.022858 141 </obs>
<obs>69200 -0.014556 142 </obs>
<obs>68200 0.011662 143 </obs>
<obs>69000 0.000000 144 </obs>
<obs>69000 0.005780 145 </obs>
<obs>69400 0.012885 146 </obs>
<obs>70300 -0.010007 147 </obs>
<obs>69600 0.002869 148 </obs>
<obs>69800 -0.021724 149 </obs>
<obs>68300 0.007294 150 </obs>
<obs>68800 0.002903 151 </obs>
<obs>69000 -0.002903 152 </obs>
<obs>68800 -0.030999 153 </obs>
<obs>66700 -0.007524 154 </obs>
<obs>66200 0.022406 155 </obs>
<obs>67700 -0.026948 156 </obs>
<obs>65900 -0.001519 157 </obs>
<obs>65800 0.004549 158 </obs>
<obs>66100 0.040035 159 </obs>
<obs>68800 0.002903 160 </obs>
<obs>69000 -0.013129 161 </obs>
<obs>68100 0.005857 162 </obs>
<obs>68500 0.000000 163 </obs>
<obs>68500 0.005822 164 </obs>
<obs>68900 -0.008746 165 </obs>
<obs>68300 0.010197 166 </obs>
<obs>69000 -0.008734 167 </obs>
<obs>68400 0.000000 168 </obs>
<obs>68400 -0.014728 169 </obs>
<obs>67400 0.002963 170 </obs>
<obs>67600 0.005900 171 </obs>
<obs>68000 0.008785 172 </obs>
<obs>68600 -0.005848 173 </obs>
<obs>68200 -0.011800 174 </obs>
<obs>67400 0.005917 175 </obs>
<obs>67800 -0.004435 176 </obs>
<obs>67500 0.024873 177 </obs>
<obs>69200 0.046586 178 </obs>
<obs>72500 0.016416 179 </obs>
<obs>73700 0.004062 180 </obs>
<obs>74000 0.001350 181 </obs>
<obs>74100 0.016065 182 </obs>
<obs>75300 -0.006662 183 </obs>
<obs>74800 0.030285 184 </obs>
<obs>77100 -0.001298 185 </obs>
<obs>77000 -0.026317 186 </obs>
<obs>75000 -0.004008 187 </obs>
<obs>74700 -0.006716 188 </obs>
<obs>74200 -0.005405 189 </obs>
<obs>73800 0.002706 190 </obs>
<obs>74000 -0.014976 191 </obs>
<obs>72900 0.014976 192 </obs>
<obs>74000 0.012089 193 </obs>
<obs>74900 -0.006698 194 </obs>
<obs>74400 0.004024 195 </obs>
<obs>74700 0.004008 196 </obs>
<obs>75000 -0.020203 197 </obs>
<obs>73500 -0.002725 198 </obs>
<obs>73300 0.006798 199 </obs>
<obs>73800 0.037238 200 </obs>
<obs>76600 0.024502 201 </obs>
<obs>78500 0.049700 202 </obs>
<obs>82500 0.000000 203 </obs>
<obs>82500 -0.030772 204 </obs>
<obs>80000 0.024693 205 </obs>
<obs>82000 0.015729 206 </obs>
<obs>83300 0.077361 207 </obs>
<obs>90000 0.005540 208 </obs>
<obs>90500 0.015351 209 </obs>
<obs>91900 -0.009842 210 </obs>
<obs>91000 0.027102 211 </obs>
<obs>93500 0.001069 212 </obs>
<obs>93600 0.066140 213 </obs>
<obs>100000 -0.004008 214 </obs>
<obs>99600 0.023811 215 </obs>
<obs>102000 0.084557 216 </obs>
<obs>111000 -0.018182 217 </obs>
<obs>109000 -0.037388 218 </obs>
<obs>105000 -0.028988 219 </obs>
<obs>102000 -0.029853 220 </obs>
<obs>99000 0.044452 221 </obs>
<obs>103500 -0.024451 222 </obs>
<obs>101000 -0.035268 223 </obs>
<obs>97500 0.020305 224 </obs>
<obs>99500 -0.025447 225 </obs>
<obs>97000 -0.020834 226 </obs>
<obs>95000 0.012552 227 </obs>
<obs>96200 0.033728 228 </obs>
<obs>99500 0.034571 229 </obs>
<obs>103000 -0.039609 230 </obs>
<obs>99000 0.039609 231 </obs>
<obs>103000 -0.009756 232 </obs>
<obs>102000 -0.029853 233 </obs>
<obs>99000 0.029853 234 </obs>
<obs>102000 -0.019803 235 </obs>
<obs>100000 0.000000 236 </obs>
<obs>100000 0.029559 237 </obs>
<obs>103000 -0.019608 238 </obs>
<obs>101000 0.009852 239 </obs>
<obs>102000 0.019418 240 </obs>
<obs>104000 -0.009662 241 </obs>
<obs>103000 0.014458 242 </obs>
<obs>104500 0.108704 243 </obs>
<obs>116500 0.042023 244 </obs>
<obs>121500 0.048202 245 </obs>
<obs>127500 0.030890 246 </obs>
<obs>131500 0.018833 247 </obs>
</observations>
</gretldata>
11:23 a.m.
On Wed, 11 Jan 2006, Seung Hun Han wrote:
when i estimate the garch parameter of cj stock price during 1 year
data
using gretl and R.
IN Gnu-R.
> garch(x,c(1,1))
... [snip] ...
Note here:
***** FALSE CONVERGENCE *****
FUNCTION -0.849489E+03 RELDX 0.114E-13
FUNC. EVALS 78 GRAD. EVALS 21
PRELDF 0.477E-13 NPRELDF 0.231E-06
I FINAL X(I) D(I) G(I)
1 0.167570E-05 0.100E+01 0.188E+04
2 0.630399E-01 0.100E+01 0.113E+01
3 0.944707E+00 0.100E+01 0.364E+00
and here
Warning message:
NANs... in: sqrt(pred$e)
Note that the sum of the 2 garch parameters alpha and beta exceeds 1. In
these cases, the GARCH model is simply not appropriate (it must be said
that your sample is a bit small for garch estimation).
On the same data, Eviews gives:
C 0.001085 0.001262 0.860183 0.3897
C 3.01E-06 3.16E-06 0.953145 0.3405
ARCH(1) 0.056867 0.018513 3.071698 0.0021
GARCH(1) 0.946872 0.019763 47.91246 0.0000
and you have the same problem.
Laurent's ox package gives
Maximum Likelihood Estimation (Std.Errors based on Second derivatives)
Coefficient Std.Error t-value t-prob
Cst(M) 0.001329 0.0011434 1.163 0.2461
Cst(V) 0.016847 0.053572 0.3145 0.7534
ARCH(Alpha1) 0.062022 0.020664 3.001 0.0030
GARCH(Beta1) 0.945226 0.025128 37.62 0.0000
and you have the same problem again.
In circumstances such as these, gretl concludes there is no convergence.
Actually, after 3 iterations the gretl algorithm lands onto
Parameters and gradients at iter. 3
0.078722 (-4.495820)
0.006162 (150.511348)
0.059888 (103.997100)
0.940112 (122.291993)
and refuses to go further because following the optimisation algorithm
you'd go outside the admissible parameter space.
See also the thread on this same list about a month ago:
http://ricardo.ecn.wfu.edu/pipermail/gretl-users/2005-December/
Hope this helps,
Riccardo `Jack' Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
jack(a)dea.unian.it
http://www.econ.univpm.it/lucchetti
6814
days inactive
6814
days old
1 comments
2 participants
participants (2)
-
jack -
Seung Hun Han