The relationship between cause and effect.

learn more… | top users | synonyms (1)

1
vote
0answers
13 views

Can external cause ever be discounted?

We are often advised not to confuse correlation with causation. Fortunately techniques do exist to assess the likelihood that an outcome does indeed result from a cause. 1 2 But in some cases it is ...
1
vote
1answer
9 views

causal impact - adding multiple control groups

I want to run an analysis using causal impact tool. I have one test group but multiple control groups. Can I use multiple control groups all together in one model? Eg: Y = test and A,B,C as control ...
0
votes
1answer
22 views

What does the error “pre.period must span at least 3 time points” in the CausalImpact R package mean?

I've been encountering the error "pre.period must span at least 3 time points" when using the package. Can someone help me understand why the package requires me to have at least 3 time points and ...
7
votes
1answer
50 views

What are the pros and cons of employing LASSO for causality analysis?

It looks like social sciences are impressed by Statistical Learning and its results. A couple of months ago, I heard Imbens saying: "LASSO is the new OLS". My problem with this is that I've been ...
1
vote
1answer
19 views

Figuring out causation from event data?

Suppose I have event data showing these things: users who signed up users who signed up and then went on to do main important activity users who signed up and then went on to invite a coworker ...
1
vote
0answers
13 views

When does the Rubin Causal Model fail in practice?

The Rubin Causal Model frames the causal inference question as the problem of inferring missing potential outcomes (what the outcome would have been if a unit had received a different treatment) in ...
1
vote
2answers
40 views

Identifying a confounder

I'm trying to check whether a variable is a confounder or not. Specifically, for a randomized trial where I want to investigate the effects of a reduction in class size on student performance, would ...
3
votes
2answers
135 views

Granger causality and non-linear regression

I’m new to Granger Causality concept. I know that the “Granger causality” is a statistical concept of causality that is based on prediction. According to Granger causality, if a time series X ...
0
votes
0answers
15 views

Treament By Covariate Interactions with Propensity Score Weighting

I am trying to estimate the causal effect of a treatment on an outcome using propensity score weighting. I estimated the propensity scores and verified covariate balance with a number of covariates, ...
0
votes
0answers
18 views

Causality between non stationary time series

I have two financial time series since both are non-stationary what is the best method to calculate the causality between two non-stationary time series?
0
votes
0answers
21 views

Adjusting for confounders when the investigated exposures are gene mutations

I'm delving into causality and directed acyclic graph for choosing the right covariate structure for multivariable regression analysis. Reading Pearl work, I understood that one should adjust only ...
0
votes
0answers
18 views

Real time Causality calculation for financial time series

I have 0.2 million financial time series data (1-minute data, each minute 1 sample data point), I want to find the causality in real time like if someone give me a data point how I know that one point ...
0
votes
1answer
41 views

Causality between multiple time series

I have 1000 financial time series (closing prices), I am using Toda-Yamamoto test. It is impossible to calculate the causality manually as there are $C_{1000}^{2-1000}$ cases. Is there any way in R ...
0
votes
0answers
14 views

Causal inference designs and research questions

Consider the following two research questions from Heckman, 2006: P.1 Forecasting the impacts (constructing counterfactual states) of interventions implemented in one environment in other ...
4
votes
1answer
33 views

What is the result of violated exclusion restrictions?

I have a question regarding exclusion restrictions in instrumental variable design. If I have an instrumental variable, which is also somewhat related to the outcome, would that (and how) cause ...
0
votes
0answers
16 views

No unmeasured confounders assumption

I am looking for a graphical interpretation of the no unmeasured confounders (NUC) assumption used in the case of dynamic treatment regime (Chakraborty and Moodie, 2013, p.13). I can't see how for any ...
2
votes
1answer
34 views

Is there sense in applying causal inference methods to variables with low correlation?

This question is somehow similar to Does causation imply correlation?, but what I would like to know is there any sense in applying a causal inference methods when we have a low correlation level. I'm ...
1
vote
1answer
35 views

Can fixed effects be “post-treament”?

I have located a natural experiment in a time series cross-sectional dataset, but I am unsure of whether or not to include unit-level fixed effects in my models. I have produced a toy example that ...
0
votes
0answers
24 views

Relationship of WTI oil with selected US sectors before, during and after a crisis

My group and I want to analyze the relationship of WTI oil with selected US sectors before, during and after a crisis. We use daily data from Datastream. Our time intervals are splitted in three ...
3
votes
0answers
70 views

Reverse causality, a bigger problem than I initially thought?

Take a standard regression framework: $$ Y_{it} =\beta X_{it} + \epsilon_{it}$$ Assume for simplicity that no omitted variables exist, nor are there simultaneity or measurement problems. In several ...
0
votes
0answers
26 views

Granger causality with matrices [duplicate]

I have a short question regarding Granger causality. I understand Granger only intuitively. If I am correct, it happens when $y_{t-1}$ and $x_t$ are better in explaining $y_t$ than only $y_{t-1}$. ...
2
votes
0answers
32 views

reverse causality when dependent variable and independent variable are observed on different levels

I am trying to regress the following: $$ Y_{ijt} = \gamma \cdot P_{jt} +X_{ijt} \beta+\epsilon_{ijt} $$ The j dimension is only indicated for clarity, the regression is a panel regression on i-t ...
6
votes
1answer
145 views

The difference between average and marginal treatment effect

I have been reading some papers, and I am unclear about the specific definitions of Average Treatment Effect (ATE), and Marginal Treatment Effect (MTE). Are they the same? According to Austin... ...
0
votes
0answers
33 views

Selecting Control Time Series for CausalImpact package in R

I'm currently using the CausalImpact package in R to analyze advertising impact on single store sales. I am working with daily sales data from nearly 2,000 locations. Of the 2,000 stores, only about ...
2
votes
1answer
64 views

Causal impact response time series

I am trying to analyze the effects of an online advertising campaign. The campaign was in market globally except for "Country A". In my response time series, I am using orders from all countries ...
0
votes
0answers
28 views

Statistical analysis using small sample size N=11 or 15

I am analyzing firm level data to unpack the cost of producing a renewable energy technology. I have dependent variable as the production cost of the technology, independent variables are three ...
1
vote
1answer
50 views

Unconfoundedness in Rubin's Causal Model- Layman's explanation

When implementing Rubin's causal model, one of the (untestable) assumptions that we need is unconfoundedness, which means $$(Y(0),Y(1))\perp T|X$$ Where the LHS are the counterfactuals, the T is the ...
0
votes
0answers
12 views

Interpreting mediation analysis output

I am trying to carry out some mediation analysis for my dataset hl. Essentially, I am trying to find out the causal mediation effects of weight (wfa: corrected for age) on the total effect that a low ...
3
votes
1answer
34 views

Causality in Online Classification

I'm using an SVM for an online classification, i.e. datapoints are classified as they come in. Of course, the training has occurred earlier, offline, with an annotated dataset. However, the system ...
0
votes
0answers
22 views

How to distinguish a causal chain model from a double effect model?

I am used a linear mixed model on two datasets (two different species), with the same explanatory variable (an environmental variable) for fixed effects. I then select the top 0.1% response ...
0
votes
0answers
24 views

Exogeneity Interpretation

Consider the following linear regression model:$$Y=X'\beta+u$$ If I wish to estimate this equation by OLS, I have to first think of ways in which the estimator might be biased. More speciifcally, I ...
2
votes
1answer
15 views

Why do I still need to include control after propensity score matching?

On page 209 of Gelman and Hill book, the authors suggest that Having created and checked appropriateness of the matches by examining balance, we fit a regression model just on the matched data ...
1
vote
1answer
55 views

CausalImpact Vs Synth

Can CausalImpact package be used in lieu of the Synth package to create a synthetic control ? The R implementation of the Synth package is very confusing compared to the Stata demo for the Synth ...
5
votes
1answer
46 views

Permutation testing in multiply adjusted analyses

Has there been or is there a consensus about how permutation testing should be done in multiply adjusted regression analyses? I understand the notion of "iteratively permuting the outcome variable" so ...
0
votes
0answers
5 views

How to retrieve the group of variables most implied in duplicates?

How to retrieve the group of variables most implied in duplicates ? I believe a good start would be to retrieve the ordered list of the most together repeated groups of variables: Given a sample or ...
0
votes
1answer
88 views

CausalImpact with Panel Data

The way I understand it, the CausalImpact package estimates a Bayesian model meant to analyse time-series data. What if, instead, I want to use panel data (i.e. repeated observations of the same ...
0
votes
0answers
29 views

Estimate a curvilinear relationship (inverted U shaped) using difference-in-differences estimator

If I want to estimate a potential curvilinear relationship (inverted U shaped) between X and Y, whether it is possible to use difference-in-differences regression (suppose I can explore variations in ...
0
votes
1answer
31 views

Analyzing the impact of change in a time series on another

I have two time series data: $x_t $ and $y_t$. I have developed an algorithm to detect unusual changes (activities) on my $x_t$ series, so that whenever there is a change detected (based on some ...
2
votes
1answer
218 views

Fixed effects in regression discontinuity design

I want to do a non parametric RDD type analysis to know the impact of an intervention (a single dummy variable) on an outcome variable. I have several 'boundaries' (which are actually different ...
0
votes
0answers
23 views

How do I estimate sensitivity of two outcomes variables from a difference-in-differences setup?

Say I estimate the following multi-period DID equations separately: $\frac{Y^1_{it}}{A_{it}} = \alpha^1_i + \alpha^1_t + \tau^1 D_i\times Post + \beta^1C_{it} + \epsilon^1_{it}$ ...
1
vote
1answer
94 views

Correlation or Causality between two time series considering a sliding window

I would perform a correlation and causality analysis between two time series considering only a little window of samples. In this way I would try to find if there is a correlation or a causality ...
0
votes
0answers
18 views

Prior on effect of treatment using CausalImpact library in R

I'm using the package causalImpact in R to estimate the causal effect of an intervention in a time series. However, I have strong prior information that the effect can't be negative. How can I encode ...
0
votes
0answers
62 views

decomposing average treatment effect

Suppose I estimate a difference-in-difference (DID) model on some outcome variable Y, and say I found a statistically significant average treatment effect (ATE). Using the same DID model on another ...
0
votes
1answer
56 views

Durbin-Watson: test exogeneity

I have a time series for which I have built a linear regression, say $Y(t)=\beta X(t)$. A regression implies that $Y$ is actually a function of $X$ (that is, $Y(X)$), but not the other way around ...
1
vote
1answer
29 views

Experimental treatments assigned via email how best estimate average treatment effects?

A growing number of social experiments are conducted outside of the laboratory, and by assigning the treatment condition through emails (e.g., often the content of the email is the intervention ...
1
vote
0answers
27 views

Methods of Treatment Effect Heterogeneity Estimation (Observation Data)

Given observational data, where a "treatment" is chosen by the unit of observation or not, are there any standard methods of ascertaining not just if the treatment has effect overall (ATT), but ...
3
votes
2answers
83 views

Understanding the direction & strength of a correlation, & causal status, from a chi-squared test

I am analyzing Stack Overflow Posts. So I have a database with 1000000 questions, their current score (upvote or downvote) and a flag, that there is a source code part in the question (or not). So I ...
1
vote
0answers
27 views

Binary logistic regression - SPSS

I did some regression analysis in SPSS using two binary variables: Biomarker X (0= low levels; 1= high levels), where 0 was the reference category and Obesity (0=no; 1=yes) ''Biomarker X'' was taken ...
2
votes
1answer
110 views

Robust Coefficients For Differences in Differences

I have a panel data set which I am looking to analyze for relationships/causality using the OLS differences-in-differences method. The panel data includes multiple observations over time for various ...
0
votes
0answers
113 views

Causal Trees to Estimate Heterogenous Treatment Effects: Transformed Outcomes [Machine Learning in Python]

I am interested in using off-the-shelf tools like scikit-learn for Python to implement the Athey-Imbens recommendation for estimating treatment effect ...