Synthetic Control

Construct a Synthetic Control for treated_ticker from a weighted combination of donor_tickers (Abadie, Diamond & Hainmueller, 2010), and estimate the effect of an event/intervention as the post-treatment_period gap between treated_ticker’s actual and synthetic counterfactual return path.

When a single company or market experiences an event (an acquisition, a regulatory change, an index inclusion) and there is no single obvious comparison, the synthetic control builds one from the data: a weighted average of unaffected “donor” tickers, with weights chosen so it tracks the treated ticker as closely as possible before the event. If that pre-event fit is good, the synthetic control’s path after the event estimates what would have happened without it, and the gap is the estimated effect.

The formula is as follows:

• All series are the tickers’ column (Return by default)

Unlike difference-in-differences, this does not require the treated and comparison units to follow parallel trends, as the synthetic control is built to match the treated ticker’s own pre-event path. Because there is only one treated unit, the p-value comes from placebo tests: the procedure is rerun with each donor as if it were treated, and the treated ticker’s post/pre fit ratio is compared with theirs. Matching is done on the pre-event outcome path only, a common simplification of the original method.

For more information about the method, see the following papers:

• Abadie, A., Diamond, A., & Hainmueller, J. (2010). “Synthetic Control Methods for Comparative Case Studies: Estimating the Effect of California’s Tobacco Control Program.” Journal of the American Statistical Association, 105(490), 493-505.
• Abadie, A. (2021). “Using Synthetic Controls: Feasibility, Data Requirements, and Methodological Aspects.” Journal of Economic Literature, 59(2), 391-425.

Also known as: SCM, synthetic control method.

Synthetic Control in Python or via the MCP

Install the Finance Toolkit with:

pip install financetoolkit -U

Then call get_synthetic_control as shown below.

from financetoolkit import Toolkit

toolkit = Toolkit(
    ["BA", "AIR.PA", "LMT", "GD", "RTX", "NOC"],
    api_key="FINANCIAL_MODELING_PREP_KEY",
    start_date="2023-01-01",
    end_date="2024-12-31",
)

# Boeing after the 737 MAX 9 grounding, with the other aerospace companies as donors
toolkit.econometrics.get_synthetic_control(treated_ticker="BA", treatment_period="2024-01-08")

Or, after setting up the Finance Toolkit MCP, ask:

Which returns:

Using Your Own Data

get_synthetic_control runs on any data, not only prices: ratios, economic indicators, yield curves or your own dataset. Pass a DataFrame, a Series or a dictionary of them to Econometrics as data, with the dates in the rows or the columns, and set changes to how its changes are calculated: "percentage", "log", "difference" for rates and ratios, or None when the data are changes already. For example, for the monthly unemployment rates of the United States and Germany, where each country is a column:

from financetoolkit import Economics, Econometrics

economics = Economics(start_date="2000-01-01")

unemployment = economics.get_unemployment_rate(
    countries=["United States", "Germany"], period="monthly"
)

econometrics = Econometrics(data=unemployment, changes="difference")

Then call econometrics.get_synthetic_control with treated_ticker and treatment_period, using the column names of the data as tickers.

Parameters

get_synthetic_control accepts the following parameters.

Related Causal Inference