Finance Research Editing and Proofreading Services

The main result survives sub-periods, alternative weighting, winsorising, industry controls and two additional factor models. Twelve rows of a robustness table, all significant. And every one of them uses the same signal construction, the same sample screen and the same portfolio formation date — so what has been demonstrated is that the result is arithmetically stable, not that it is real.

We edit what finance researchers write — asset pricing and factor papers, empirical corporate finance studies, event studies and their windows, market microstructure work, portfolio and performance evaluation, banking and credit analyses, robustness and specification sections, referee responses, and job market papers, theses and practitioner reports. Our editors work on the robustness section and what it leaves untested.

The robustness section is where a finance paper is either strengthened or padded, and its failure is a table of variations that all share the paper's central choice. Robustness is only informative when it varies the thing that could be wrong. We work through these so the fragile choices are identified first — the signal definition, the sample screen, the formation and holding period, the treatment of delisting returns and micro-caps — since these are where a result usually lives or dies; so each robustness test says which assumption it relaxes, because a table of alternative specifications with no stated purpose reads as volume; so the tests that materially weaken the result are reported with their magnitudes, given that a finding which halves out of sample is more useful stated honestly than defended; so multiple testing and the number of specifications examined are acknowledged, as a search across many signals produces significant results by construction; so transaction costs, capacity and implementability are addressed where the paper implies a strategy; so the economic magnitude is stated alongside significance, since a two-basis-point effect with a t-statistic of three is a statistical fact rather than an economic one; and so what the robustness section does not test is stated. Papers written this way are the ones that replicate.

Everything you send is treated in confidence, including working papers, proprietary data descriptions and unpublished results. We are editors rather than finance academics or analysts, and we offer no view on models, results or any investment implication. What we can do is make the robustness section informative.