Skip to content
Licensed Unlicensed Requires Authentication Published by De Gruyter March 15, 2013

Approximate p-Values of Certain Tests Involving Hypotheses About Multiple Breaks

  • Alastair Hall EMAIL logo and Nikolaos Sakkas

Abstract

We provide formulae for calculating approximate p-values for the non-standard asymptotic null distributions of a variety of tests used for detecting multiple structural change in a wide range of models. Our approximations are based on simulated quantiles obtained from 100,000 replications, and the latter are more accurate than the quantiles reported in the literature by increasing the number of replications by a factor of 10. The p-value response surfaces are approximated using a parametric method proposed by Hansen and their use is illustrated with an example. Using our p-value response surfaces, it is shown that the use of Bai and Perron’s response surfaces for the critical values of these tests can lead to misleading inferences, and thus should be used with extreme caution.


Corresponding author: Alastair Hall, Economics, School of Social Sciences, University of Manchester, Oxford Road, Manchester M13 9PL, UK

Appendix

Table 1

Absolute Errors in Fitted Distributions.

p-ValueSupFF(1+l|l)UDmaxWDmax
MedianMaximumMedianMaximumMedianMaximumMedianMaximum
0.000.00010.00030.00020.00070.00020.00070.00010.0001
0.010.00030.00120.00080.00330.00110.00320.00140.0033
0.020.00040.00150.00090.00370.00130.00380.00160.0047
0.030.00040.00140.00090.00370.00140.00340.00160.0043
0.040.00040.00150.00080.00380.00140.00320.00160.0041
0.050.00040.00150.00080.00340.00140.00310.00150.0045
0.060.00040.00150.00070.00350.00130.00280.00120.0038
0.070.00040.00140.00070.00360.00110.00270.00110.0047
0.080.00040.00140.00060.00300.00100.00230.00120.0040
0.090.00040.00140.00070.00330.00070.00200.00100.0041
0.100.00040.00130.00060.00260.00050.00200.00100.0041
0.150.00030.00140.00070.00330.00070.00250.00130.0038
0.200.00040.00150.00070.00420.00110.00270.00130.0033
0.250.00040.00170.00060.00350.00130.00320.00140.0043
0.300.00040.00160.00060.00340.00100.00310.00120.0048
0.400.00040.00180.00070.00350.00050.00140.00130.0045
0.500.00040.00170.00080.00360.00120.00350.00190.0044
0.600.00050.00160.00080.00360.00130.00420.00160.0055
0.700.00050.00210.00090.00430.00060.00190.00140.0059
Table 2a

SupF Distributions, k=1.

qε=0.05ε=0.10ε=0.15ε=0.20ε=0.25
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–1.411.083.76–1.171.053.34–1.011.023.02–0.861.002.78–0.740.982.55
2–2.341.115.53–1.991.064.94–1.701.034.56–1.441.024.36–1.260.994.01
3–3.271.136.94–2.661.106.56–2.221.076.25–1.841.066.05–1.581.045.67
4–4.081.168.46–3.291.148.21–2.771.117.81–2.451.077.22–2.081.056.87
5–4.861.149.44–4.101.118.88–3.551.078.39–3.001.058.11–2.501.047.91
6–5.811.1410.25–4.831.119.96–3.891.1210.34–3.201.1210.25–2.651.109.95
7–6.301.1511.80–5.261.1111.20–4.281.1111.47–3.661.0810.98–3.151.0510.43
8–7.231.1412.42–5.941.1312.38–4.971.1112.28–4.261.0811.86–3.741.0511.28
9–8.071.1413.19–6.571.1313.43–5.671.1012.97–4.631.1113.42–4.151.0612.45
10–8.901.1113.42–7.511.1013.54–6.321.0913.78–5.321.0813.78–4.241.0814.05
11–9.361.1515.51–7.821.1315.34–6.451.1315.79–5.571.0915.07–4.731.0714.82
12–10.561.0914.42–8.841.0814.85–7.731.0514.43–6.601.0514.69–5.651.0314.53
13–10.831.1416.97–9.071.1116.79–8.121.0615.61–6.71.0616.26–5.511.0716.73
14–8.361.1317.72–9.681.0917.30–8.361.0616.90–7.361.0416.61–5.821.0818.08
15–12.531.1117.47–10.741.0717.00–9.231.0517.07–7.821.0617.84–6.601.0618.11
16–13.261.0817.40–11.611.0115.94–10.001.0317.20–8.471.0518.24–7.421.0217.63
17–13.901.0818.34–11.721.0718.92–10.101.0619.08–8.461.0720.14–7.271.0620.17
18–14.831.0718.44–12.611.0318.35–11.061.0218.60–9.381.0319.61–7.701.0520.93
19–15.331.0619.24–13.131.0218.70–11.461.0219.50–9.621.0420.96–8.551.0220.29
20–16.161.0619.56–13.551.0721.37–11.531.0822.45–10.021.0722.78–8.501.0723.09
Table 2b

SupF Distributions, k=2.

qε=0.05ε=0.10ε=0.15ε=0.20ε=0.25
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–5.192.328.09–3.062.197.31–2.082.096.34–1.591.955.11–1.191.904.24
2–7.922.3210.62–5.172.2210.06–3.472.189.73–2.462.128.81–1.792.027.46
3–10.352.2011.49–6.892.1711.98–4.602.1912.43–3.402.1111.30–2.382.0710.28
4–12.342.1913.24–8.502.1914.25–5.622.2415.43–3.772.2014.91–2.782.0912.98
5–14.082.1814.97–9.952.1615.86–7.102.1416.31–4.772.1716.89–2.572.2117.17
6–16.012.2517.50–11.222.2519.09–8.152.2019.18–6.182.1217.95–4.012.1217.73
7–17.492.2619.74–12.512.1820.12–9.002.1821.15–7.322.0318.57–5.281.9917.73
8–19.362.1419.18–14.352.0319.00–10.652.0620.80–8.132.0420.70–5.941.9919.74
9–20.972.1921.63–15.022.1924.05–11.022.1925.31–7.482.2226.81–4.762.1926.22
10–22.722.1321.82–16.632.1624.91–11.652.2428.59–8.272.2028.64–4.312.2630.40
11–24.322.0621.59–17.632.1827.35–13.922.0425.41–10.732.0225.77–7.272.0426.71
12–25.922.1324.56–19.662.0224.80–15.561.9624.97–12.491.9525.16–8.541.9927.12
13–27.502.1626.68–20.872.1328.83–15.702.1331.02–11.132.1633.37–8.482.0731.16
14–28.852.0926.65–22.032.0227.74–17.062.0430.36–11.722.1735.69–8.082.1335.10
15–30.752.0025.10–22.712.1833.97–16.262.2538.83–11.332.2440.42–7.952.1638.59
16–31.862.0227.00–24.202.0732.19–18.862.0233.34–14.372.0335.12–10.292.0235.71
17–33.452.0328.50–24.732.1737.13–19.482.0836.74–14.052.1440.56–10.052.1040.26
18–34.581.9226.70–26.762.0333.95–20.392.1039.08–15.112.1041.06–12.181.9937.85
19–35.691.8927.16–28.221.9031.12–21.652.0338.46–15.842.0842.44–12.172.0240.98
20–37.701.9730.37–29.172.0336.76–20.792.2046.98–16.722.0944.26–13.212.0041.81
Table 2c

SupF Distributions, k=3.

qε=0.05ε=0.10ε=0.15ε=0.20
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–6.013.3711.95–3.083.2911.04–2.253.048.42–1.692.886.23
2–10.043.2914.94–5.923.2514.81–3.623.2013.75–2.822.9810.39
3–13.073.3118.34–8.243.1617.51–5.443.1016.57–3.283.1215.25
4–15.853.2320.46–9.573.3623.32–5.203.4324.37–4.263.1018.50
5–17.633.4726.97–11.023.4027.55–8.593.0522.23–5.803.0020.29
6–21.003.3727.66–12.913.4130.90–9.573.1626.85–7.422.9321.97
7–23.073.3530.45–14.933.3332.65–10.333.2031.02–7.023.0827.87
8–25.423.4334.56–17.153.2133.40–11.163.2234.69–9.462.8826.80
9–26.613.6442.26–17.723.3439.40–12.353.1937.39–8.893.0633.48
10–30.663.2335.25–21.123.0936.01–13.583.2140.68–11.552.8531.47
11–33.113.1335.05–22.513.1640.16–14.863.1642.64–9.983.0940.48
12–35.163.1537.93–24.333.1342.05–17.593.0642.00–10.763.1644.73
13–37.663.2441.97–24.763.3350.20–18.033.2048.38–9.263.3553.53
14–39.573.2544.82–25.363.4155.92–18.693.2151.96–8.593.4259.52
15–41.973.1744.49–28.173.2753.94–18.833.2857.51–9.063.3962.63
16–43.782.8638.05–30.573.1552.74–19.183.3162.00–15.343.0252.19
17–45.723.1749.32–33.332.9548.46–25.992.8246.76–14.633.1559.31
18–48.012.9243.42–33.083.2260.34–26.042.9553.67–17.453.0357.25
19–49.732.8442.89–35.893.0355.61–23.203.2067.17–19.522.8754.79
20–52.173.0450.67–36.453.1863.72–23.873.2471.46–13.673.3175.81
Table 2d

SupF Distributions, k=4.

qε=0.05ε=0.15ε=0.15
θ0θ1ηθ0θ1ηθ0θ1η
1–6.314.6017.10–2.954.3814.24–2.473.969.47
2–11.334.4621.22–6.684.1618.05–4.114.0915.44
3–15.324.3624.62–9.134.0922.23–6.014.0119.41
4–18.804.2827.77–10.694.3429.45–6.624.2126.15
5–20.524.6337.24–11.444.5137.17–8.664.1029.04
6–23.684.7142.45–13.324.5342.05–10.883.9931.20
7–27.904.4040.53–16.214.4244.12–10.934.1838.75
8–29.744.6549.46–18.194.3947.86–10.754.3045.93
9–29.934.9862.12–18.784.4454.12–12.204.2449.06
10–36.154.4452.23–23.304.1450.30–16.364.0046.50
11–39.964.1548.54–24.714.2656.73–14.814.2056.58
12–41.454.4960.50–25.944.3362.50–16.704.2360.30
13–43.964.6267.11–27.974.3867.17–17.264.3066.46
14–47.614.3362.45–27.014.5978.56–19.524.1666.58
15–51.324.0657.25–33.354.1266.77–19.184.2974.64
16–53.714.0158.80–32.384.4180.22–17.284.4886.24
17–56.403.9659.85–36.954.1574.35–26.753.9268.48
18–58.884.0164.08–39.134.1176.36–27.053.9974.65
19–59.114.4282.05–37.294.3188.96–23.234.2389.69
20–63.314.2578.24–37.904.4899.04–27.584.0886.01
Table 2e

SupF Distributions, k=5.

qε=0.05ε=0.10ε=0.15
θ0θ1ηθ0θ1ηθ0θ1η
1–6.955.4919.54–3.305.2615.53–2.204.788.86
2–11.845.5827.13–7.245.0620.65–3.724.8314.82
3–15.495.7835.52–8.975.1828.19–5.264.7919.76
4–21.315.1933.28–11.355.3135.00–5.165.1328.41
5–21.945.8749.15–11.505.6145.64–6.295.1634.00
6–25.605.8854.75–14.335.5249.78–9.834.7233.02
7–31.255.4451.16–16.615.5054.99–11.174.7938.12
8–32.695.7963.43–20.665.2354.53–9.925.0648.27
9–27.666.7093.08–18.945.5668.47–10.305.1655.24
10–39.505.6770.78–23.405.3167.01–16.064.5947.78
11–45.005.2965.11–24.985.3773.57–14.844.9059.03
12–46.845.4975.01–30.095.0569.06–13.155.1770.88
13–45.996.0696.55–29.425.4283.89–12.585.3280.00
14–51.465.7591.70–25.365.83103.75–14.715.1680.98
15–57.685.2178.73–27.355.72106.59–16.665.1083.84
16–58.945.4992.09–34.385.4098.93–19.914.9783.72
17–64.705.0981.92–36.915.40102.89–22.894.8684.01
18–68.184.8477.74–42.065.1597.81–25.464.8085.73
19–66.715.56107.76–38.655.39114.30–21.035.02100.86
20–70.065.56111.62–37.125.68131.05–26.944.8094.89
Table 2f

SupF Distributions, k=6.

qε=0.05ε=0.10
θ0θ1ηθ0θ1η
1–7.246.4722.42–3.996.0015.46
2–12.226.6932.68–7.246.0823.63
3–15.577.0144.17–9.096.1431.90
4–22.756.3141.55–11.706.2839.65
5–22.387.0860.94–10.246.8155.14
6–28.256.7561.49–17.166.1350.21
7–32.946.6765.53–17.116.4962.97
8–34.166.9979.02–21.066.2264.02
9–32.097.60101.85–18.416.6981.74
10–42.456.8387.30–24.256.3378.78
11–49.156.3780.66–26.286.3384.79
12–47.087.05106.03–32.025.9880.48
13–44.987.53128.72–28.866.54102.30
14–50.667.33127.99–25.116.85120.57
15–62.216.33100.46–30.006.59118.34
16–61.316.94126.03–36.856.29112.43
17–69.626.35110.21–38.426.36120.20
18–73.516.20110.01–43.536.15117.34
19–67.097.13152.80–35.196.73149.05
20–69.977.19160.97–31.197.04170.03
Table 2g

SupF Distributions. k=7.

qε=0.05ε=0.10
θ0θ1ηθ0θ1η
1–8.087.1723.23–4.086.9816.07
2–12.317.7837.87–7.496.9824.68
3–17.167.8047.07–9.567.0533.69
4–24.427.1646.28–11.287.3844.50
5–22.198.3373.08–9.587.9761.55
6–28.408.0175.22–16.427.3258.77
7–30.848.2287.47–19.007.2364.75
8–34.868.1694.27–18.417.5378.38
9–32.958.74117.45–18.697.6789.63
10–45.237.83100.34–26.617.0581.84
11–52.677.3593.96–28.987.0888.74
12–43.708.66140.63–36.276.5781.03
13–45.388.74152.33–30.747.40110.29
14–53.948.33145.31–22.898.00139.26
15–69.566.85103.40–32.727.35125.13
16–65.727.95145.04–34.567.47134.78
17–71.207.73143.63–19.468.49182.98
18–78.587.21131.01–38.017.55151.04
19–51.719.31230.80–33.837.79170.45
20–71.628.46195.76–28.678.17195.04
Table 2h

SupF Distributions. k=8.

qε=0.05ε=0.10
θ0θ1ηθ0θ1η
1–8.747.9524.30–3.598.0915.84
2–12.368.8542.54–7.117.8223.81
3–17.038.9454.17–8.908.0334.12
4–24.558.3655.55–11.178.1843.54
5–22.799.4281.86–9.778.7359.77
6–29.479.0784.82–18.107.7253.23
7–33.219.0393.61–16.508.3370.78
8–37.749.00101.32–21.747.8469.96
9–38.069.39120.01–19.728.3187.41
10–50.948.31100.42–23.518.1691.53
11–53.528.62115.01–31.347.5083.27
12–43.609.83161.54–29.508.09103.61
13–36.2510.55197.66–24.078.73129.72
14–56.519.36163.33–27.568.42130.58
15–71.508.16130.66–43.057.1298.98
16–69.348.94163.69–40.807.68121.29
17–66.959.38189.36–37.248.16144.16
18–73.459.17188.75–22.739.12190.43
19–59.5210.14243.14–34.958.46172.41
20–66.7010.01244.4512.3410.57283.52
Table 2i

SupF Distributions, k=9.

qε=0.05
θ0θ1η
1–9.328.6925.10
2–12.209.9347.07
3–14.5710.5066.24
4–24.919.4262.67
5–22.7010.5691.44
6–29.6410.1695.11
7–30.0410.53114.25
8–38.8910.00112.09
9–37.5610.56136.33
10–53.149.23111.33
11–54.899.61129.09
12–34.6911.59202.38
13–45.5211.06195.16
14–55.0410.61189.53
15–71.789.44157.89
16–74.819.69173.25
17–60.4510.94233.31
18–67.0310.79236.23
19–47.3511.87301.11
20–43.7112.18328.79
Table 3a

FT(+1|) Distributions, =1.

qε=0.05ε=0.10ε=0.15ε=0.20ε=0.25
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–2.41.174.8–2.001.144.44–1.751.114.06–1.481.113.93–1.281.103.72
2–3.821.257.07–3.241.206.54–2.781.196.35–2.441.186.10–2.161.155.67
3–5.431.207.63–4.611.207.61–3.921.217.78–3.371.227.83–2.941.207.54
4–6.721.259.21–5.821.228.92–5.131.198.60–4.471.188.54–3.911.188.46
5–7.821.2310.20–6.851.189.54–6.111.169.27–5.291.199.77–4.561.189.81
6–9.151.1710.02–8.061.149.69–7.161.1710.33–6.461.1610.19–5.761.149.85
7–10.101.2312.06–8.761.2111.98–7.811.2012.01–6.921.1912.04–6.161.1711.79
8–11.331.1511.18–10.061.1411.38–8.931.1411.74–8.011.1411.79–7.131.1512.10
9–12.321.2113.25–10.911.1612.55–9.801.1613.00–8.961.1412.63–7.971.1412.74
10–13.291.1613.07–11.921.1112.30–10.811.1212.82–9.681.1413.57–8.511.1714.53
11–14.551.1713.76–12.831.1614.18–11.551.1614.60–10.181.2015.93–9.291.1515.04
12–15.341.1013.07–13.651.1013.70–12.431.0913.77–11.291.0913.99–10.401.0413.10
13–16.441.1615.04–14.561.1515.55–13.201.1215.14–11.921.1416.16–10.871.1215.95
14–17.291.1515.70–15.421.1215.69–13.851.1517.12–12.621.1417.10–11.671.1116.41
15–18.351.1215.45–16.361.1115.77–14.821.1216.75–13.621.1016.50–12.531.0615.73
16–19.021.0915.46–16.991.1217.21–15.641.0214.69–14.521.0214.86–13.301.0114.92
17–20.001.1116.50–17.921.0716.31–16.401.0516.08–15.141.0416.07–13.921.0617.11
18–20.951.0916.36–18.761.1318.63–17.041.1319.41–15.671.0918.60–14.441.0617.96
19–21.591.0616.18–19.321.1219.30–17.841.0517.42–16.481.0216.92–15.291.0016.64
20–22.571.0516.40–20.451.0216.39–18.891.0216.92–17.531.0117.10–16.221.0317.91
Table 3b

FT(+1|) Distributions, =2.

qε=0.05ε=0.10ε=0.15ε=0.20
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–3.151.225.34–2.661.205.08–2.311.184.82–2.001.184.69
2–5.051.307.52–4.301.277.25–3.761.277.18–3.361.256.86
3–6.901.227.66–6.041.227.69–5.371.227.75–4.791.237.88
4–8.481.248.86–7.541.228.55–6.741.228.75–6.081.228.72
5–9.731.239.78–8.641.229.70–7.831.209.55–7.061.2310.11
6–11.111.2210.32–9.951.169.61–9.161.1810.01–8.451.149.42
7–12.441.3012.76–11.101.2512.01–10.181.2211.56–9.301.1911.25
8–13.501.2011.70–12.161.1210.44–11.231.1410.93–10.341.1210.77
9–14.991.2312.91–13.431.1712.05–12.371.1912.57–11.381.1411.83
10–15.931.1912.77–14.401.1512.33–13.301.1211.90–12.341.1512.89
11–17.251.1412.12–15.661.1713.34–14.471.1713.63–13.321.2014.78
12–18.091.1413.07–16.391.1112.93–15.231.1413.9013.991.1815.27
13–19.301.1513.83–17.591.1313.80–16.221.1214.19–15.161.1314.55
14–20.271.1213.76–18.511.1214.29–17.241.1515.56–16.031.1415.68
15–21.451.1214.21–19.501.1114.77–18.151.1315.69–16.931.1215.79
16–22.291.1215.08–20.471.1416.20–18.701.0815.05–17.471.0715.14
17–23.401.1215.42–21.361.1216.25–19.691.0815.81–18.331.0816.04
18–24.341.1015.29–22.561.1317.16–20.961.1317.70–19.421.1318.29
19–24.911.0715.11–23.261.1217.50–21.401.0616.22–19.961.0817.53
20–25.701.0414.72–23.611.0114.70–22.081.0014.81–20.981.0215.70
Table 3c

FT(+1|) Distributions, =3.

qε=0.05ε=0.10ε=0.15
θ0θ1ηθ0θ1ηθ0θ1η
1–3.771.265.81–3.221.255.60–2.721.275.67
2–6.051.317.60–5.211.307.49–4.641.307.50
3–7.951.237.68–7.121.217.39–6.451.207.40
4–9.631.238.47–8.681.198.03–7.941.238.58
5–11.111.2810.17–9.981.229.42–9.111.219.41
6–12.591.2710.81–11.291.199.77–10.561.209.96
7–14.281.3412.76–12.791.2611.74–11.821.2311.30
8–15.301.3113.34–13.731.2011.58–12.751.1510.82
9–16.821.2211.98–15.181.1911.81–14.171.1912.01
10–17.731.1912.20–16.181.1712.38–15.071.1612.40
11–18.791.1110.91–17.581.1612.46–16.411.1612.90
12–19.831.1312.12–18.321.1412.89–17.241.1613.56
13–21.371.1613.40–19.631.1313.19–18.181.1113.12
14–22.091.1012.50–20.401.1013.13–19.371.1213.89
15–23.331.1113.19–21.631.1213.97–20.381.1314.70
16–24.511.1314.50–22.821.1515.71–21.041.1215.36
17–25.571.1114.18–23.781.1315.48–22.111.1215.84
18–26.501.0914.36–25.001.1215.81–23.551.1416.78
19–27.031.0614.03–25.701.1116.12–23.761.0615.33
20–28.001.0514.26–25.851.0314.35–24.201.0013.94
Table 3d

FT(+1|) Distributions, =4.

qε=0.05ε=0.10ε=0.15
θ0θ1ηθ0θ1ηθ0θ1η
1–4.331.296.07–3.731.295.94–3.241.305.99
2–6.851.337.64–6.001.317.57–5.411.317.46
3–8.751.227.48–7.891.207.30–7.261.217.34
4–10.561.299.20–9.471.177.67–8.851.218.24
5–12.241.3010.32–11.021.249.47–10.111.229.40
6–13.821.3311.70–12.381.2210.07–11.641.229.99
7–15.711.3512.60–14.081.2711.52–13.071.2411.23
8–16.801.3213.04–15.121.2612.46–13.991.2211.84
9–18.031.2011.07–16.491.1911.41–15.481.1911.56
10–19.061.1811.53–17.511.1812.13–16.541.2112.87
11–19.801.0910.26–18.881.1411.63–17.841.1512.04
12–21.201.1311.90–19.791.1412.46–18.751.1613.11
13–22.891.1712.98–21.131.1412.80–19.551.0912.25
14–23.411.0911.88–21.851.1012.63–20.701.0912.49
15–24.651.1012.58–23.251.1313.66–21.991.1213.76
16–26.191.1413.98–24.631.1615.20–22.931.1515.54
17–27.021.0913.20–25.481.1214.49–23.971.1215.16
18–28.101.0913.72–26.581.1014.46–25.381.1315.59
19–28.421.0413.07–27.361.1015.07–25.431.0514.35
20–29.771.0613.99–27.761.0514.45–25.791.0013.32
Table 3e

FT(+1|) Distributions, =5.

qε=0.05ε=0.10
θ0θ1ηθ0θ1η
1–4.821.316.22–4.171.316.18
2–7.501.337.61–6.641.327.58
3–9.361.217.31–8.521.207.15
4–11.391.349.82–10.181.228.24
5–13.191.3210.39–11.861.259.49
6–14.921.4012.62–13.341.2810.79
7–16.851.3512.17–15.121.2811.35
8–18.001.3212.74–16.261.2812.42
9–18.911.1810.45–17.461.1710.73
10–20.051.1711.05–18.661.1912.01
11–20.401.069.60–19.861.1411.25
12–22.381.1411.78–20.851.1311.78
13–23.941.1512.30–22.261.1312.23
14–24.461.0911.54–23.101.1112.53
15–25.891.1112.43–24.541.1313.37
16–27.521.1413.59–25.971.1514.37
17–28.091.0812.54–26.731.1113.67
18–29.181.0812.85–27.621.0813.34
19–29.531.0412.37–28.521.0814.04
20–31.171.0713.71–29.221.0614.13
Table 3f

FT(+1|) Distributions, =6.

qε=0.05ε=0.10
θ0θ1ηθ0θ1η
1–5.241.326.31–4.571.336.36
2–8.061.327.47–7.201.327.56
3–9.821.197.08–9.021.197.05
4–12.161.369.88–10.841.268.70
5–13.991.3310.29–12.561.269.50
6–15.931.4212.70–14.201.3311.50
7–17.791.3411.67–16.021.2911.39
8–18.931.3112.21–17.241.2912.32
9–19.631.169.92–18.171.1510.12
10–20.891.1710.86–19.601.1911.77
11–21.161.079.58–20.601.1310.82
12–23.281.1411.40–21.761.1311.52
13–24.701.1411.65–23.101.1211.68
14–25.431.1011.46–24.261.1312.71
15–26.921.1212.26–25.521.1212.84
16–28.681.1513.42–27.141.1614.07
17–28.781.0611.78–27.681.1013.05
18–29.891.0611.98–28.481.0712.64
19–30.351.0211.75–29.231.0512.89
20–32.211.0713.19–30.421.0713.89
Table 3g

FT(+1|) Distributions, =7.

qε=0.05ε=0.10
θ0θ1ηθ0θ1η
1–5.631.316.26–4.961.336.36
2–8.521.327.41–7.681.327.44
3–10.251.207.14–9.411.176.84
4–12.831.379.93–11.431.278.77
5–14.661.3210.00–13.151.269.39
6–16.811.4312.51–15.021.3812.31
7–18.551.3311.17–16.801.3011.35
8–19.701.3011.66–18.071.2912.10
9–20.271.169.62–18.791.149.83
10–21.581.1710.63–20.391.2011.59
11–22.141.1010.11–21.151.1110.40
12–24.111.1511.28–22.511.1211.13
13–25.261.1211.11–23.621.0910.92
14–26.201.1011.24–25.261.1512.81
15–27.861.1312.23–26.351.1212.45
16–29.601.1513.09–28.041.1513.61
17–29.181.0410.98–28.381.0812.41
18–30.661.0611.66–29.191.0612.10
19–31.091.0211.40–29.761.0311.96
20–32.981.0612.70–31.411.0713.62
Table 3h

FT(+1|) Distributions, =8.

qε=0.05
θ0θ1η
1–5.991.296.08
2–8.931.327.35
3–10.661.207.13
4–13.441.399.99
5–15.241.329.79
6–17.581.4412.41
7–19.171.3110.74
8–20.391.3011.37
9–20.811.159.40
10–22.001.1510.11
11–23.091.1410.67
12–24.831.1511.18
13–25.671.1010.55
14–27.121.1211.48
15–28.731.1412.25
16–30.331.1512.67
17–29.481.0210.37
18–31.181.0511.20
19–31.681.0211.10
20–33.621.0512.29
Table 4a

UDmax Distributions.

qε=0.05ε=0.10ε=0.15ε=0.20ε=0.25
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–4.731.454.36–3.041.344.06–2.121.263.76–1.491.183.39–1.061.112.97
2–6.711.344.86–4.681.244.73–3.471.184.62–2.561.144.52–1.851.094.28
3–8.191.234.99–5.961.185.27–4.551.155.51–3.441.125.59–2.521.085.47
4–9.561.205.42–7.201.155.85–5.591.136.24–4.321.106.36–3.221.086.45
5–10.681.155.72–8.211.116.18–6.481.076.45–5.101.066.87–3.851.077.30
6–11.891.136.03–9.281.096.66–7.481.087.22–5.941.077.76–4.521.068.17
7–12.881.106.32–10.051.056.89–8.181.057.64–6.551.038.10–5.041.028.60
8–13.801.056.22–11.011.037.10–9.011.017.79–7.341.018.49–5.741.029.27
9–14.981.066.81–11.941.037.69–9.871.028.58–8.121.039.46–6.361.0210.16
10–15.691.016.59–12.801.007.72–10.691.018.93–8.771.019.87–6.801.0311.18
11–16.641.006.91–13.600.998.27–11.411.019.73–9.371.0110.69–7.361.0211.76
12–17.360.966.75–14.320.968.08–12.110.959.10–10.230.9610.03–8.220.9811.43
13–18.440.977.26–15.160.958.32–12.830.949.27–10.780.9610.85–8.690.9812.25
14–18.920.937.14–15.630.918.29–13.440.939.87–11.390.9410.90–9.141.0013.50
15–19.890.937.36–16.620.928.71–14.330.9510.55–12.100.9511.79–9.940.9613.18
16–20.300.907.22–17.200.908.85–14.870.9110.16–12.710.9412.03–10.480.9413.24
17–21.440.917.67–18.010.909.34–15.430.8910.27–13.260.9212.07–11.000.9514.11
18–22.110.897.70–18.640.889.28–16.190.8910.64–13.930.9112.45–11.490.9514.95
19–22.160.857.25–19.020.858.84–16.830.8911.09–14.490.9213.21–12.130.9214.59
20–23.560.877.96–20.000.869.50–17.610.9011.89–15.260.9314.00–12.640.9616.38
Table 4b

WDmax Distributions.

qε=0.05ε=0.10ε=0.15ε=0.20ε=0.25
θ0θ1ηθ0θ1ηθ0θ1ηθ0θ1ηθ0θ1η
1–3.151.455.06–2.471.384.59–2.111.314.12–1.531.243.71–1.141.193.29
2–5.051.346.03–4.041.376.36–3.331.336.15–2.511.305.86–1.841.175.02
3–6.701.286.57–5.491.236.39–4.801.246.48–3.551.196.42–2.071.267.56
4–8.351.439.06–6.871.388.89–5.741.449.97–4.761.278.00–3.011.268.70
5–9.551.379.72–8.131.217.92–7.321.258.52–5.181.3010.05–3.721.259.77
6–10.831.3810.91–8.951.3411.06–7.861.2810.38–6.441.158.93–4.631.129.26
7–12.141.5013.93–10.141.4012.99–9.251.3111.56–6.251.3714.13–3.721.3514.93
8–13.191.3612.65–11.301.3713.34–9.581.4615.75–7.451.3314.19–3.301.4318.48
9–14.661.4114.21–12.511.4114.98–10.751.4616.70–6.991.4318.27–5.541.3016.13
10–15.121.139.96–13.521.1911.55–11.981.3014.56–9.081.2615.01–4.751.3719.82
11–17.201.3113.71–14.251.4618.58–12.781.4418.51–9.101.3718.97–5.371.3520.70
12–16.901.0910.47–15.041.1312.03–13.481.2014.31–10.711.1314.22–8.721.0613.41
13–19.451.2914.70–17.231.2514.64–15.321.2114.52–12.291.0812.87–8.481.2719.48
14–19.701.2615.46–16.811.2115.71–15.131.2016.14–12.681.0713.90–9.591.1016.29
15–21.151.2816.23–17.791.2316.90–16.531.2016.24–12.371.2620.32–9.851.1518.54
16–22.331.3017.45–19.821.2316.50–17.981.3520.84–14.641.1918.14–11.071.1619.25
17–23.551.2516.65–20.901.2718.31–19.061.2919.78–15.091.2420.41–11.721.1620.01
18–25.011.2617.13–21.751.1816.44–20.211.2619.49–14.201.4228.49–12.531.1720.98
19–24.561.2318.38–21.761.1015.48–20.601.1417.03–16.771.0516.31–13.650.9715.47
20–26.561.2017.01–23.301.1215.96–21.751.1818.40–17.871.0817.49–14.141.1120.23
Table 5

p-Value accuracy comparison of Hall and Sakkas (2011) (HS) and Bai and Perron (2003b) (BP).

kq=1q=5q=10q=15q=20
BPHSBPHSBPHSBPHSBPHS
10.03830.04850.05400.04940.05260.04970.07680.04960.16200.0492
20.05240.04900.06530.04990.06220.04940.10960.04870.29930.0490
30.04870.04940.06080.04950.05620.04910.11300.04940.35880.0497
40.05770.04970.06240.04990.05320.04920.11560.04960.40590.0495
50.07670.04940.06840.04960.05270.04950.11980.05030.44600.0497
60.11610.04960.08150.04970.05500.04930.12710.05080.48760.0495
70.19120.04980.10340.04970.06100.05010.13880.05070.53050.0502
80.32690.05040.13830.04990.07140.05030.15430.05010.57670.0498
90.54190.05010.19130.05010.08610.05050.17510.04970.62420.0494

The authors acknowledge the support of the ESRC grant RES-062-23-1351. We would like to thank Jonathan Boyle and Robin Pinning of the University of Manchester Research Computing Services for their assistance in using Condor. We are also grateful to Denise Osborn for useful comments on various aspects of the work. This paper was presented at the 4th CSDA international conference on computational and financial econometrics (CFE ’10), 10-12 December 2010, London, UK.

  1. 1

    The error must also satisfy certain other conditions detailed in the sources mentioned below.

  2. 2

    If p=0 then all parameters are regime specific and the model is said to exhibit pure structural change.

  3. 3

    It should be noted that these limiting distributions only apply for the statistics based on FT(λ1,…,λk;q) if the regression error ut is homoscedastic and serially uncorrelated. If the errors are heteroscedastic and/or serially correlated then the same limiting distributions hold for the analogous functions of Wald statistics that satisfactorily account for the error structure; see Bai and Perron (1998) for further details.

  4. 4

    Such large values of ε are reported because Andrews (1993) allows for asymmetric trimming.

  5. 5

    This methodology has also been employed by Sen and Hall (1999) for the distributions of their overidentifying restrictions test (OT) and by Sen (1999) for the Ghysels, Guay, and Hall’s (1997) predictive test.

  6. 6

    From the case of q=1, it can be seen that this case occurs for k=9. Further investigation reveals the main source of the distortion is Bai and Perron’s (2003b) response surface: the appropriate critical value reported in Bai and Perron (1998) is 5.20 which translates, using our response surfaces, to a p-value of 0.0758; the critical value predicted by Bai and Perron’s (2003b) response surface is 3.8023. In contrast, our simulations (based on an enhanced design) yield a critical value of 5.4210 for which our response surface gives a p-value of 0.05006.

  7. 7

    In the original version of this paper, we restricted attention to q=1,5,10 but, given the nature of the results for those cases, a referee conjectured Bai and Perron’s (2003b) response surfaces would work well for cases q>10. While natural given results for q=1, 5, 10, this conjecture is erroneous as our results demonstrate. We therefore included these cases to avoid the reader making the same mistake. We attribute the inaccuracy of Bai and Perron’s (2003b) response surfaces to these cases to the fact that the surfaces were calculated using cases for which q≤10 only.

  8. 8

    For completeness, we note that we performed similar calculations for ε=0.10,0.15,0.20,0.25 for q=1,5,10. The distortions are smaller than those reported for ε=0.05 but are nevertheless non-trivial in a number of circumstances. These results are available from the authors upon request.

References

Andrews, D. W. K. 1993. “Tests for Parameter Instability and Structural Change with Unknown Change Point.” Econometrica 61: 821–856.10.2307/2951764Search in Google Scholar

Andrews, D. W. K. 2003. “Tests for Parameter Instability and Structural Change with Unknown Change Point: A Corrigendum,” Econometrica 71: 395–398.10.1111/1468-0262.00405Search in Google Scholar

Bai, J., H. Chen, T. Chong, and S. Wang. 2008. “Generic Consistency of the Breakpoint Estimators Under Specification Errors in a Multiple-Break Model.” Econometrics Journal 11: 287–307.10.1111/j.1368-423X.2008.00237.xSearch in Google Scholar

Bai, J. and P. Perron. 1998. “Estimating and Testing Linear Models with Multiple Structural Changes.” Econometrica 66: 47–78.10.2307/2998540Search in Google Scholar

Bai, J. and P. Perron. 2003a. “Computation and Analysis of Multiple Structural Change Models.” Journal of Applied Econometrics 18: 1–22.10.1002/jae.659Search in Google Scholar

Bai, J. and P. Perron. 2003b. “Critical Values for Multiple Structural Change Tests.” Econometrics Journal 6: 72–78.10.1111/1368-423X.00102Search in Google Scholar

Boldea, O. and A. R. Hall. 2013. “Estimation and Inference in Unstable Nonlinear Least Squares Models.” Journal of Econometrics 172: 158–167.10.1016/j.jeconom.2012.09.004Search in Google Scholar

Ghysels, E., A. Guay, and A. R. Hall. 1997. “Predictive Test for Structural Change with Unknown Breakpoint.” Journal of Econometrics 82: 209–233.10.1016/S0304-4076(97)00057-2Search in Google Scholar

Hall, A. R., S. Han, and O. Boldea. 2012. “Inference Regarding Multiple Structural Changes in Linear Models with Endogenous Regressors.” Journal of Econometrics 170: 281–302.10.1016/j.jeconom.2012.05.006Search in Google Scholar PubMed PubMed Central

Hansen, B. E. 1997. “Approximate Asymptotic p Values for Structural-Change Tests.” Journal of Business and Economic Statistics 15: 60–67.10.2307/1392074Search in Google Scholar

Sen, A. 1999. “Approximate p-Values of Predictive Tests for Structural Stability.” Economics Letters 63: 245–253.10.1016/S0165-1765(99)00031-2Search in Google Scholar

Sen, A. and A. R. Hall. 1999. “Two Further Aspects of Some New Tests for Structural Stability.” Structural Change and Economic Dynamics 10: 431–443.10.1016/S0954-349X(99)00011-9Search in Google Scholar

Published Online: 2013-03-15
Published in Print: 2013-07-01

©2013 by Walter de Gruyter Berlin Boston

Downloaded on 23.4.2024 from https://www.degruyter.com/document/doi/10.1515/jem-2012-0014/html
Scroll to top button