Endogeneity Analysis without the access of raw data?

Without access to the raw data, conducting a comprehensive analysis of endogeneity becomes challenging, as many traditional methods rely on directly manipulating and analyzing the data. However, there are still some approaches you can consider:

Instrumental Variables (IV) Analysis: If you have access to instrumental variables that are correlated with the endogenous variable but not directly related to the outcome, you may be able to conduct an IV analysis. IV methods can help address endogeneity by providing unbiased estimates of causal effects even in the presence of omitted variables or measurement error. However, identifying valid instruments without access to the raw data can be difficult.

Granger Causality Test: If you have time series data, you can use the Granger causality test to examine whether one variable "Granger-causes" another variable. This test can help identify potential endogeneity issues by assessing the direction of causality between variables.

Sensitivity Analysis: Conduct sensitivity analyses to assess the robustness of your regression results to potential endogeneity. This involves testing the impact of alternative model specifications, variable transformations, or estimation techniques on your findings.

Review Existing Literature: Review existing literature on the subject area to identify common sources of endogeneity and potential strategies for addressing them. This can help inform your analysis and interpretation of regression results.

Expert Consultation: Consult with experts in the field who may have access to similar data or experience dealing with endogeneity issues in similar contexts. They may be able to provide valuable insights or suggest alternative approaches.

While conducting a comprehensive analysis of endogeneity without access to raw data is challenging, these approaches may help you gain some insights into the potential presence of endogeneity and its implications for your regression results.

Please follow me if it's helpful. All the very best. Regards, Safiul

Harshavardhana Anantharaman Srinivasan

As a provisional solution, I have employed Cholesky decomposition to simulate data, which can generate a correlation matrix that closely approximates the original. Nevertheless, this method presents potential issues because it constructs data predicated on the assumption that all features are normally distributed. Please advise whether this is an appropriate course of action.

Mohammad Mamun

Conducting an endogeneity analysis without direct access to raw data presents challenges but is feasible with certain limitations. One approach involves utilizing instrumental variable (IV) techniques or natural experiments if suitable instruments or exogenous shocks are available. IV methods require identifying variables that affect the endogenous variable but are unrelated to the outcome except through their effect on the endogenous variable. However, without access to raw data, identifying appropriate instruments becomes difficult. Alternatively, leveraging pre-existing research, meta-analyses, or publicly available datasets may offer insights into potential instruments or exogenous shocks. Additionally, employing robustness checks, sensitivity analyses, or econometric techniques like difference-in-differences can help mitigate endogeneity concerns when direct access to raw data is unavailable. Nevertheless, the validity and reliability of such analyses depend on the quality and relevance of the data and methods employed.

Mohammad Mamun

Endogeneity analysis typically requires access to raw data or at least detailed information about the variables of interest and their relationships. However, conducting endogeneity analysis without access to raw data can be challenging and may limit the depth of the analysis. Here are some approaches you can consider if you don't have access to raw data:

Literature Review: Start by reviewing existing literature on the topic of interest. Look for studies that have addressed endogeneity issues similar to yours and examine their methodologies, including how they handled endogeneity concerns. This can provide insights into potential strategies or techniques you can apply in your analysis.

Theoretical Considerations: Based on your understanding of the subject matter and theoretical framework, try to identify potential sources of endogeneity in your analysis. Consider factors that could lead to correlation or causality issues between variables and think about how these issues might be addressed or mitigated.

Instrumental Variables: If you have access to instrumental variables that are plausibly exogenous and relevant to your analysis, you can use them to address endogeneity concerns. Instrumental variables estimation requires careful selection and validation of instruments, so make sure to justify their relevance and validity in your context.

Quasi-Experimental Methods: Explore quasi-experimental methods or natural experiments that exploit exogenous variation in the data. These methods can help identify causal effects while controlling for endogeneity. Examples include difference-in-differences, regression discontinuity design, and propensity score matching.

Sensitivity Analysis: Perform sensitivity analysis to assess the robustness of your results to potential sources of endogeneity. This involves testing the sensitivity of your findings to different model specifications, control variables, and assumptions. Sensitivity analysis can provide insights into the reliability and stability of your results.

Expert Consultation: If possible, consult with experts or researchers who have experience with the data or topic area. They may offer valuable insights or suggest alternative approaches to address endogeneity concerns given the limitations of the available data.

While conducting endogeneity analysis without access to raw data presents challenges, it's still possible to employ various strategies and techniques to mitigate endogeneity concerns and produce meaningful results. However, it's essential to acknowledge the limitations of the analysis and interpret the findings cautiously, considering the potential impact of unobserved factors and data limitations on the results.

Is there any selective media to grow filamentous fungus only (not with yeast)?

Do anyone know a NK cell line expressing FC receptor to perform ADCC assay?

I want to simulate a VAR model. How do I choose the coefficients to ensure stability?

What is the pebbling number of a cycle graph with 4 vertices?

Anyone know where the micro tensile testing machine available?

How do I etch nichrome thin film on polyimide(kapton) substrate? Can someone explain the procedure?

Is there an association between non-vegetarianism and violence?

What could be the possible reasons for getting 20 unit less value of mass of pure compound which is well charecterised by NMR?

Different significance values on SPSS and PLS SEM. Please help.

Is gender a dichotomous variable? If yes, Can t-test be run on single large sample dichotomous variable (demographic)?

How to compare two groups with only two measurements?

Can we classify HV and LV lines based on line length?

What's the best way to measure growth rates in House sparrow chicks from day 2 to day 10?

What analysis to choose with a large neste dataset and clearly skewed or kurtotic distributions?

How to calculate average trend in maximum annual NDVI from MODIS?

New Journal, "Integrative Psychological and Behavioral Science" -- do you not know, and have you not seen, this done before?

Which is a lucid standard comprehensive book to guide on psychological scale construction and its standardization?

Do you know any MR image dataset?

Can NFL theorem be valid in infinite search space in ML?

Does anyone know HRM channel in a 2 plex QIAGEN Rotor- Gene Q qPCR can be used as a separate channel?