ALL Metrics
-
Views
-
Downloads
Get PDF
Get XML
Cite
Export
Track
Research Article

Informative prior on structural equation modelling with non-homogenous error structure

[version 1; peer review: 1 approved, 1 approved with reservations]
PUBLISHED 04 May 2022
Author details Author details
OPEN PEER REVIEW
REVIEWER STATUS

Abstract

Introduction: This study investigates the impact of informative prior on Bayesian structural equation model (BSEM) with heteroscedastic error structure. A major drawback of homogeneous error structure is that, in most studies the underlying assumption of equal variance across observation is often unrealistic, hence the need to consider the non-homogenous error structure.
Methods: Updating appropriate informative prior, four different forms of heteroscedastic error structures were considered at sample sizes 50, 100, 200 and 500.
Results: The results show that both posterior predictive probability (PPP) and log likelihood are influenced by the sample size and the prior information, hence the model with the linear form of error structure is the best.
Conclusions: The study has been able to address sufficiently the problem of heteroscedasticity of known form using four different heteroscedastic conditions, the linear form outperformed other forms of heteroscedastic error structure thus can accommodate any form of data that violates the homogenous variance assumption by updating appropriate informative prior. Thus, this approach provides an alternative approach to the existing classical method which depends solely on the sample information.

Keywords

Bayesian SEM, Latent Variable, Observed Variable, Heteroscedastic error structure, Predictive Performance

Introduction

Bayesian structural equation modeling (BSEM) analyses the relationship between the observed, unobserved, and latent variables within the Bayesian context.14,16,21,24 The data visualization can be done by path diagram. In Bayesian inference, θ is random, which depicts the level of uncertainty about the true value of θ because both the observed data y and the parameters θ are assumed random. The joint probability of the parameters and the data as functions of the conditional distribution of the data given the parameters, and the prior distribution of the parameters can be modelled. More formally,

(1)
p(θY)pθp(Yθ)

where

P(θ|y) is the posterior distribution

P(θ) is the prior distribution

P(y|θ) is the likelihood function

The un-normalized posterior distribution when expressed in terms of the unknown parameters θ for fixed values of y, this term is the likelihood L(θ|y). Thus, can be rewritten as:

(2)
p(θY)pθL(θy)

Studies abound on classical methods and Bayesian methods with a focus on homogeneous variance.8,19,22,25 This study explores the BSEM using different forms of heteroscedastic error structure.

Methods

Bayesian estimation of structural equation models (SEM)

This section develops a Gibbs sampler to estimate SEM with reflective measurement indicators.1,11,12 The Bayesian estimation is illustrated by considering a SEM that is equivalent to the mostly used model. A SEM is composed of a measurement equation (3) and a structural equation (4)9:

(3)
yi=Λωi+εi
(4)
ηi=Πηi+Γξi+δi
where iϵ1n

It is assumed that measurement errors are uncorrelated with ω and δ, residuals are uncorrelated with ω and the variables are distributed as follows:

(5)
εiN0Ψε
(6)
δiN0Ψδ
(7)
ωiN0Σω

iϵ1n, where Ψε and Ψδ are diagonal matrices. The covariance matrix of ω is derived based on the SEM:

(8)
Σω=EηηTEξηTEηξTEξξT
(9)
Σω=Π01ΓΦΓT+ΨδΠ0TΠ01ΓΦΦΓTΠ0TΦ
(10)
ηηT=Π01Γξ+Π01δΠ01Γξ+Π01δT=Π01ΓξξTΓT+δδTΠ0T+Π01ΓξδT+δξTΓTΠ0TEηηT=Π01ΓΦΓT+ΨδΠ0TηξT=Π01Γξ+Π01δξTEηξT=Π01Γ

Prior distributions

In order to enable Gibbs sampling from full conditional posterior distributions, natural conjugate prior distributions for the unknown parameters are considered.25 Let ψεκ be the kth diagonal element of Ψε, ψδι be the lth diagonal element of Ψδ,ΛκT be the kth row of Λ and MιT be the lth row of M,

(11)
ψεk1Gammaα0εk,β0εk
(12)
Λkψεk1NΛ0kψεkH0Λk
(13)
ψδi1Gammaα0δi,β0δi
(14)
Miψδi1NM0iψδiH0Mi
(15)
ΦIWv0V0
with κϵ1p and ιϵ1q1

Derivations of conditional distributions

The joint posterior of all unknown parameters is proportional to the likelihood times the prior,

(16)
pΛΨεΩMΨδΦYpYΛΨεΩMΨδΦpΛΛεΩMΨδΦ

Given Y and Ω, Λ and Ψε are independent from Σω. Draws of Ω, can cause estimation of Λ and Ψε as a simple regression model. Thus, sampling from the posterior distribution of Λ and Ψε without reference to Σω. The same holds for inference with regard to M, Φ and Ψδ, which are independent from Y given Ω.

Heteroscedastic error structures

The heteroscedastic error structure with different functional form of error variance under consideration are double logarithmic form, linear form, linear-inverse form and linear-absolute form as expressed in equation 17, 18, 19 and 20, respectively.

(17)
σ2=lnσ2=λo+λilnγi+vi
(18)
σ2=|εiεi'|=λi+λ2γi+νi2
(19)
σ2=|εiεi'|=λi+λ2γi+νi2
(20)
σ2=|εiεi'|=λi+λ2γi+νi2

Each of the functional forms of heteroscedastic error structure will be incorporated into the modified model. The variance matrix for disturbance vector is given as

(21)
=εiεj'=σλ2ii=j
(22)
Ω=σ2λi000σ2λ200000σ2λn

The posterior distribution

The posterior density is the product of the likelihood and the prior distribution chosen2,13

(23)
(Pλh,Ω|y)αpyλhΩpλphpΩ
(24)
pPλhΩy=hN2exph2γλ'γλγ×hN+vk2exphv2s2×pΩ×exp12λλ0V1¯λλ0
(25)
PλhΩy=hN2|Ω|12exph2yλγΩ1yλγ×expv1¯2λ¯λ0V1¯λ¯λ0×n1α+1expβh×Ω1β0+k+1/2etrR01β01/2

Since the full posterior distribution is intractable; a Markov chain Monte Carlo (MCMC) simulation method of Gibbs sampling is employed.25 This involves the use of marginal posterior distribution.

(26)
λ=λ0=γΩ1γ1γ'Ω1γ=γ'γ1γ'γ
S2=γγλ0γγλ0V¯

Also

Vs2¯+λλ^0'γi'γiλλ^0=γγiλ'γγiλ
(27)
pγiγiσ2=hv+k/22πN2exp(h2v¯S2+λλ^0'γi'γiλλ^0
v¯=NKandN=v¯+K

Consider an informative prior created by set.

v1¯j=1ckjγjγj

And letting c 0forj=1,2

The posterior distribution of λ conditional on γ, h, Ω is given by:

(28)
pλγhΩαhN2exp(h2γλγ'Ω1γλγ+λλ0'V1¯γλ0×exph2Ωγiλγi'γiλγi+λλ0'λλ0v¯

Solving the exponential part of the above equation, we will have:

γiλγi'γiλγi=γ2+λγ22yiλγandγλ0'λλ0=λ2+λ022λλ0

Therefore,

=exph2Ωi=1Nyi2+λγ22yiλγi+λλ0v¯2

The additional term not involving λ is factored out to give:

(29)
=expλ22v2¯+λ2λ0v2¯+λnyΩσ2nλ22Ωσ2

Factorization in terms of λ, the term in the exponential becomes:

=λ2σ2+2λλ2σ2
σ2=1v¯+nΩσ21andλ=σ2λ02v¯+nyΩσ2

So, the posterior density of λ conditioned on other parameter h, Ω, y is a multivariate normal with mean λ and variance σ2.

That is,

pλhΩyNλσ2

The posterior distribution of h conditional on λ, Ω, y is given by:

(30)
PλhΩγαhN2exph2γλγ'Ω1γλγ×hN+vk2exphv2S2=hN+vk2exph2ΩiNyi2+nλγ22ynλγhv2s2

The posterior distribution of Ω*, conditional on y*, λ*, h, is given by:

(31)
PΩyλhαPΩ×hN2exph2yλγΩ1yλγ
(32)
PΩyλhαhN2exph2yλγΩ1yλγ×|Ω|(β0+k+1)/2exptrR01β01/2

The Gibbs sampler

The Gibbs sampling procedure used in this study involves generation of sequence of draws from the conditional posterior distribution of each parameter.2,22,25

Gibbs sampling procedure

  • (i) Chose a starting or initial value, ϕ0 for s=1,2,,S

  • (ii) Take a random draw, ϕ1x from the full conditional, pϕ1yϕ1x1

  • (iii) Take a random draw, ϕ2x from the full conditional, pϕ2yϕ1x using the updated values of ϕ1x

  • (iv) Repeat until M draws are obtained, each being a vector of ϕx

  • (v) Perform the Burn-in by dropping the first S0 of these draws to eliminate the effect of ϕ0, the remaining S1 draws are then averaged to obtain the estimate of the posterior Egϕ/y.

The right-hand side of (15) is proportional to the density function of an inverse Wishart distribution

Then,

(33)
PΦYΩIWqΩΩTR01n+ρ0

Design of simulation

  • At different functional forms of3 heteroscedastic error structure with changes in sample size of 50, 100, 200 and 500. Hyper-parameter will be arbitrarily chosen for the simulation using Gibbs sampler an MCMC method.6,22

  • The R code can be accessed via the Extended data.26

  • Factor loading and error precision followed multivariate normal and inverse gamma distributions respectively to assess the prior sensitivity.21

  • The criteria that will be used to assess the performance of the posterior simulation technique are the posterior estimates.

In order to evaluate the Bayesian model fit, we used the posterior predictive probability (PPP) procedure.4,5,7,24

(34)
PPP=P(fyλ̂i<fyrepλ̂i1mi=1mδi

After achieving convergence (after j iterations). λ̂j+1λj+1Ωj+1 can be regarded as observation from p(λ*, Ω|y) collect λtΩtt=j+1.+T for statistical inference.

(35)
λ̂=T1t=1Tλt,Ω̂=T1t=1TΩt
gives Bayesian estimates of parameter and the latent variables.10,17,23

Results and discussion

The section presents the discussion of analysis of results; performances of the estimators across the parameters for the different forms of heteroscedasticity, performances of Bayesian posterior simulation and analytical methods in the presence of heteroscedasticity via consideration of four (4) different forms of heteroscedastic error structures over four sample sizes of 50, 100, 200 and 500.

Performance of the estimators at heteroscedasticity condition

This gives the results for the latent and observed variables at various sample sizes for the four heteroscedastic error conditions considered.

Comparison of latent variable estimates at different sample sizes under the heteroscedasticity condition

Using the assumed values for the estimates which are λ1 = 2.0, λ2 = 3.0 and precision = 15.0.

The covariance matrix of ω was derived to be EηξT=01Γ with M at fixed values (0 or 1). The Bayesian estimates of SEM using the independent normal-gamma priors were derived for the two classes of SEM. Hyper-parameter was arbitrarily chosen for the simulation using Gibbs sampler a Markov chain Monte Carlo (MCMC) method since the joint posterior density does not have a tractable form. For the double logarithmic form, at 95% credible interval, when n=50, Posterior Mean, PM, and Precision, PR (2.011, 2.435, and 13.202), Posterior Standard Deviation PSD (0.035, 0.033, and 0.223) and when n=100, PM, and PR (2.022, 2.528, and 13.70), PSD (0.023, 0.025, and 0.251), when n=200, PM, and PR (2.052, 2.611, and 14.4), PSD (0.017, 0.018, and 0.255), when n=500, PM, and PR (2.010, 2.801, and 14.7), PSD (0.031, 0.021, and 0.258).

For the linear form, when n=50, PM, and PR (1.845, 2.779, and 13.95), PSD (0.240, 0.242, and 0.235). When n=100, PM, and PR (1.861, 2.811, and 14.22), PSD (0.328, 0.226, and 0.325), when n= 200, PM, and PR (1.956, 2.921, and 14.72), PSD (0.219, 0.217, and 0.212), and when n=500, PM, and PR (2.120, 3.122, and 14.95), PSD (0.211, 0.311, and 0.114).

For the linear-inverse form when n=50, PM, and PR (1.882, 2.742, and 14.95), PSD (0.040, 0.028, and 0.291). When n=100, PM, and PR (1.972, 2.835, and 14.65), PSD (0.024, 0.023, and 0.229). When n=200, PM, and PR (1.988, 2.901, and 14.45), PSD (0.017, 0.016, and 0.109), and when n=500, PM, and PR (2.021, 3.003, and 14.21), PSD (0.011, 0.015, and 0.105).

For the linear-absolute form, when n=50, PM, and PR (2.036, 2.824, and 14.500), PSD (0.032, 0.034, and 0.122), When n=100, PM, and PR (1.908, 2.903, and 13.92), PSD (0.022, 0.026, and 0.234). When n=200, PM, and PR (1.893, 2.809, and 13.85), PSD (0.017, 0.023, and 0.311), and when n=500, PM, and PR (1.806, 2.788, and 13.55), PSD (0.031, 0.035, and 0.433).

Examining different forms of heteroscedastic error structures in Bayesian structural equation modeling using informative priors, rather than assuming homogenous variance which is often a statistical fallacy in many studies. We compare the models’ posterior means and standard deviations in Tables 1, 2, 3 and 4. The differences are unlikely to impact substantive conclusions, but two of them are noteworthy.

Table 1. Double logarithmic form on latent variable and observed variable estimates.

Sample sizesLatent variablesPosterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)Measured variablesEstimateStandard Deviation
n=50λ12.0110.0351.9592.062x10.0450.023
λ22.4350.0332.3842.485x20.0380.023
Precision (PR)13.2020.22313.07113.332
n=100λ12.0220.0231.9792.064x10.0530.008
λ22.5280.0252.4842.571x20.0370.024
Precision13.7000.25113.56113.838
N=200λ12.0520.0172.0152.088x10.0060.045
λ22.6110.0182.5732.648x20.0480.020
Precision14.40.25514.26014.539
N=500λ12.0100.0311.9612.058x10.0400.028
λ22.8010.0212.7602.841x20.0180.004
Precision14.70.25814.55914.840

Table 2. Linear form on latent variable and observed variable estimates.

Sample sizesLatent variablesPosterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)Measured variablesEstimateStandard Deviation
n=50λ11.8450.2401.7091.981x10.0780.017
λ22.7790.2422.6432.915x20.0550.036
Precision13.9500.23513.81614.844
n=100λ11.8610.3281.7022.0197x10.0790.012
λ22.8110.2262.6792.943x20.0360.028
Precision14.2200.32514.06214.378
N=200λ11.9560.2191.8262.086x10.0710.008
λ22.9210.2172.7923.050x20.0470.016
Precision14.720.21214.54214.898
N=500λ12.1200.2111.9932.247x10.0520.022
λ23.1220.3112.9673.277x20.0590.010
Precision14.950.11414.85715.044

Table 3. Linear inverse form on latent variable and observed variable estimates.

Sample sizesLatent variablesPosterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)Measured variablesEstimateStandard Deviation
n=50λ11.8820.0431.8271.937x10.0750.020
λ22.7420.0282.6962.788x20.0230.017
Precision14.950.29114.80115.099
n=100λ11.9720.0241.9292.015x10.0550.010
λ22.8350.0232.7932.877x20.0310.021
Precision14.650.22914.31714.583
N=200λ11.9880.0171.8262.102x10.0540.006
λ22.9010.0162.7903.012x20.0320.024
Precision14.450.10914.35814.541
N=500λ12.0210.0111.9922.050x10.0520.015
λ23.0030.0152.9693.037x20.0500.022
Precision14.2100.10514.12014.300

Table 4. Linear absolute form on latent variable and observed variable estimates.

Sample sizesLatent variablesPosterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)Measured variablesEstimateStandard Deviation
n=50λ12.0360.0321.9862.086x10.0430.018
λ22.8240.0342.7732.875x20.0270.022
Precision14.5000.12214.40314.597
n=100λ11.9080.0221.8671.949x10.0470.017
λ22.9030.0262.8582.948x20.0430.025
Precision13.920.23413.78614.054
N=200λ11.8930.0171.8571.929x10.0540.017
λ22.8090.0232.7672.851x20.0410.024
Precision13.850.31113.69614.005
N=500λ11.8060.0311.7571.855x10.0480.019
λ22.7880.0352.7362.840x20.0440.022
Precision13.550.43313.36713.732

First, the posterior means of the loadings ( λ1 and λ2 ) are somewhat smaller under different heteroscedastic condition with the informative priors as observed in Tables 6 and 7. Second, the factor variance γ is larger under our model with informative priors, likely because the informative prior placed more density on larger values of the posterior standard deviation. An evaluation of the model fit was based on the values of PPP as shown in Table 5 and it was observed that the linear form is the best with minimum PPP value as sample size increases. It was also revealed by the downward slope of the model as the sample size increases from 50 to 500 shown in Figure 1b when compared with Figure 1a, 2a and 2b.

Table 5. Comparison at varying sample sizes of different heteroscedastic form.

Sample sizeDouble logarithmicLinearLinear inverseLinear absolute
LogLikPPPLogLikPPPLogLikPPPLogLikPPP
N=50-17.5770.538-17.3090.501-19.7010.567-20.0650.560
N=100-24.3240.543-43.0580.523-16.2140.544-19.7770.544
N=200-29.4270.541-44.9350.545-15.3050.540-19.5470.532
N=500-35.5100.482-60.9200.570-14.4940.531-18.1710.506

Table 6. Latent variable estimates at different sample sizes under the double-logarithmic and linear forms.

Sample sizeLatent variablesDouble logarithmicLinear
Posterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)Posterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)
N=50λ12.0010.2311.8682.1342.1100.2301.9772.243
λ22.2830.5382.0802.4862.5540.2012.4302.678
N=100λ12.0210.3121.8662.1762.0200.1231.9232.117
λ22.4780.5622.2702.6862.6010.3562.4362.766
N=200λ12.0320.4321.8502.2142.0110.1741.8952.127
λ22.7700.8322.5173.0232.7050.4562.5182.892
N=500λ12.1000.4451.9152.2852.0050.2531.8662.144
λ22.8881.5642.5413.2343.1020.5752.8923.312

Table 7. Latent variable estimates at different sample sizes under the linear-inverse and linear absolute forms.

Sample sizeLatent variablesLinear-inverseLinear-absolute
Posterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)Posterior Mean (PM)Posterior Standard Deviation (PSD)Credible Interval (CI)
N=50λ12.1010.3521.9372.2651.7320.3111.5771.887
λ22.6370.5282.4362.8382.5820.5832.3702.794
N=100λ11.9820.4211.8022.1621.8100.2521.6711.949
λ22.7540.1922.6332.8752.6340.3752.4642.804
N=200λ11.9750.4761.7842.1661.8200.2111.6961.947
λ22.8140.9012.5513.0772.7230.7662.4802.966
N=500λ12.1110.4881.9172.3051.9200.1451.8152.026
λ23.0731.1022.7823.3642.9020.3312.7433.062
be145828-c12f-4eb6-ba82-111f3c0b680e_figure1.gif

Figure 1. Plot of log likelihood and posterior predictive probability (PPP) at various sample sizes under (a) the double logarithmic form and (b) the linear form.

be145828-c12f-4eb6-ba82-111f3c0b680e_figure2.gif

Figure 2. Plot of log likelihood and posterior predictive distribution (PPP) at various sample sizes under (a) the linear-inverse form (b) the linear-absolute form.

Considering an improvement to maximum likelihood method, in Bayesian estimations, parameters are considered as random with informative prior distribution also known as the conjugate family of the posterior, once the data is simulated/collected, it is combined with prior distribution using Bayes theorem, next posterior distribution is calculated reflecting the prior knowledge and simulated data.14,15,21 Joint posterior distribution is summarized using MCMC simulation techniques in terms of lower dimensional summary statistics as posterior mean and posterior standard deviations.5,25 We observe that the structural and measurement equation obtained from this study are adequate and in general we could accept the proposed model.

Conclusion

In this research, the derived Bayesian estimators of a structural equation model in the presence of different forms of heteroscedastic error structures validated accurate statistical inference. The study has also been able to address sufficiently the problem of heteroscedasticity of known form using four different heteroscedastic conditions for both linear and quadratic forms, and it has also successfully modified the homogenous error structure to heteroscedastic error structure in Bayesian structural equation model.20 The linear form outperformed other forms of heteroscedastic error structure thus can accommodate any form of data that violates the homogenous variance assumption by updating appropriate informative prior.16,18 Thus, this approach provides an alternative approach to the existing classical method which depends solely on the sample information.

Data availability

Underlying data

All data underlying the results are available as part of the article and no additional source data are required.

Extended data

Figshare: RCODE BSEM.docx. https://doi.org/10.6084/m9.figshare.19299851.26

Data are available under the terms of the Creative Commons Zero “No rights reserved” data waiver (CC0 1.0 Public domain dedication).

Comments on this article Comments (0)

Version 2
VERSION 2 PUBLISHED 04 May 2022
Comment
Author details Author details
Competing interests
Grant information
Copyright
Download
 
Export To
metrics
Views Downloads
F1000Research - -
PubMed Central
Data from PMC are received and updated monthly.
- -
Citations
CITE
how to cite this article
Olalude OA, Muse BO and Alaba OO. Informative prior on structural equation modelling with non-homogenous error structure [version 1; peer review: 1 approved, 1 approved with reservations]. F1000Research 2022, 11:494 (https://doi.org/10.12688/f1000research.108886.1)
NOTE: If applicable, it is important to ensure the information in square brackets after the title is included in all citations of this article.
track
receive updates on this article
Track an article to receive email alerts on any updates to this article.

Open Peer Review

Current Reviewer Status: ?
Key to Reviewer Statuses VIEW
ApprovedThe paper is scientifically sound in its current form and only minor, if any, improvements are suggested
Approved with reservations A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.
Not approvedFundamental flaws in the paper seriously undermine the findings and conclusions
Version 1
VERSION 1
PUBLISHED 04 May 2022
Views
14
Cite
Reviewer Report 26 Jul 2022
Mohamed R. Abonazel, Department of Applied Statistics and Econometrics, Faculty of Graduate Studies for Statistical Research, Cairo University, Giza, Egypt 
Approved with Reservations
VIEWS 14
This paper investigated the impact of informative prior on Bayesian structural equation model with heteroscedastic error structure. Four different forms of heteroscedastic error structures were considered. The suggested Bayesian approach provides an alternative approach to the existing classical method which ... Continue reading
CITE
CITE
HOW TO CITE THIS REPORT
Abonazel MR. Reviewer Report For: Informative prior on structural equation modelling with non-homogenous error structure [version 1; peer review: 1 approved, 1 approved with reservations]. F1000Research 2022, 11:494 (https://doi.org/10.5256/f1000research.120326.r136840)
NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article.
Views
12
Cite
Reviewer Report 12 Jul 2022
Adenike Oluwafunmilola Olubiyi, Department of Statistics, Ekiti State University, Ado Ekiti, Nigeria 
Approved
VIEWS 12
ABSTRACT
  1. In this paper, the research team investigates the impact of informative prior on Bayesian Structural equation model (BSEM) with heteroscedastic error structure.
     
  2. The drawback of homogeneous error structure was addressed
... Continue reading
CITE
CITE
HOW TO CITE THIS REPORT
Olubiyi AO. Reviewer Report For: Informative prior on structural equation modelling with non-homogenous error structure [version 1; peer review: 1 approved, 1 approved with reservations]. F1000Research 2022, 11:494 (https://doi.org/10.5256/f1000research.120326.r140694)
NOTE: it is important to ensure the information in square brackets after the title is included in all citations of this article.

Comments on this article Comments (0)

Version 2
VERSION 2 PUBLISHED 04 May 2022
Comment
Alongside their report, reviewers assign a status to the article:
Approved - the paper is scientifically sound in its current form and only minor, if any, improvements are suggested
Approved with reservations - A number of small changes, sometimes more significant revisions are required to address specific details and improve the papers academic merit.
Not approved - fundamental flaws in the paper seriously undermine the findings and conclusions
Sign In
If you've forgotten your password, please enter your email address below and we'll send you instructions on how to reset your password.

The email address should be the one you originally registered with F1000.

Email address not valid, please try again

You registered with F1000 via Google, so we cannot reset your password.

To sign in, please click here.

If you still need help with your Google account password, please click here.

You registered with F1000 via Facebook, so we cannot reset your password.

To sign in, please click here.

If you still need help with your Facebook account password, please click here.

Code not correct, please try again
Email us for further assistance.
Server error, please try again.