Hi,

I am currently writing my thesis, using panel data to assess the relationship between financial development and economic growth. I want to control for alleged endogeneity, in the financial development indicators, I have used xtivreg2, but the Hansen J Stat is significant, which states that that the instruments are invalid. What would be suggested to do now?

This is the code I used:

xtivreg2 loggdppercap GrosscapitalformationofGD Schoolenrollmentsecondary (StockmarketcapitalizationtoG LiquidliabilitiestoGDPG = l.StockmarketcapitalizationtoG l2.StockmarketcapitalizationtoG l.LiquidliabilitiestoGDPG l2.LiquidliabilitiestoGDPG ) , fe cluster(c_id) endog( LiquidliabilitiestoGDPG StockmarketcapitalizationtoG)

Results:


------------------------
Number of groups = 89 Obs per group: min = 3
avg = 18.1
max = 28

IV (2SLS) estimation
--------------------

Estimates efficient for homoskedasticity only
Statistics robust to heteroskedasticity and clustering on c_id

Number of clusters (c_id) = 89 Number of obs = 1610
F( 4, 88) = 16.32
Prob > F = 0.0000
Total (centered) SS = 51.85112797 Centered R2 = 0.3885
Total (uncentered) SS = 51.85112797 Uncentered R2 = 0.3885
Residual SS = 31.70731729 Root MSE = .1444

----------------------------------------------------------------------------------------------
| Robust
loggdppercap | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-----------------------------+----------------------------------------------------------------
StockmarketcapitalizationtoG | .000945 .0005407 1.75 0.080 -.0001147 .0020047
LiquidliabilitiestoGDPG | .000902 .0005874 1.54 0.125 -.0002492 .0020533
GrosscapitalformationofGD | .0047421 .0028177 1.68 0.092 -.0007804 .0102646
Schoolenrollmentsecondary | .0092139 .0014798 6.23 0.000 .0063135 .0121143
----------------------------------------------------------------------------------------------
Underidentification test (Kleibergen-Paap rk LM statistic): 18.271
Chi-sq(3) P-val = 0.0004
------------------------------------------------------------------------------
Weak identification test (Cragg-Donald Wald F statistic): 1543.425
(Kleibergen-Paap rk Wald F statistic): 2035.606
Stock-Yogo weak ID test critical values: 5% maximal IV relative bias 11.04
10% maximal IV relative bias 7.56
20% maximal IV relative bias 5.57
30% maximal IV relative bias 4.73
10% maximal IV size 16.87
15% maximal IV size 9.93
20% maximal IV size 7.54
25% maximal IV size 6.28
Source: Stock-Yogo (2005). Reproduced by permission.
NB: Critical values are for Cragg-Donald F statistic and i.i.d. errors.
------------------------------------------------------------------------------
Hansen J statistic (overidentification test of all instruments): 9.791
Chi-sq(2) P-val = 0.0075
-endog- option:
Endogeneity test of endogenous regressors: 12.028
Chi-sq(2) P-val = 0.0024
Regressors tested: LiquidliabilitiestoGDPG StockmarketcapitalizationtoG
------------------------------------------------------------------------------
Instrumented: StockmarketcapitalizationtoG LiquidliabilitiestoGDPG
Included instruments: GrosscapitalformationofGD Schoolenrollmentsecondary
Excluded instruments: L.StockmarketcapitalizationtoG
L2.StockmarketcapitalizationtoG L.LiquidliabilitiestoGDPG
L2.LiquidliabilitiestoGDPG

Any help is much appreciated!