A fit to five activity indices gives 59.8 to 112.7 percent of added 42.38-day amplitudes on 47 later HD 26965 nights

We test how an early activity fit changes a coherent signal in later HD 26965 velocities. Public NEID tables give 157 exposures on 83 nights. The first 36 nights determine the activity fit. The next 47 nights measure the added signal, with a free offset in each later season. The main test gives equal weight to each night. We add a 42.38-day sine at the original exposure times and hold measured activity indices fixed. We repeat the early fit after each addition. We subtract the later fitted sine and cosine coefficient vectors, then measure the length of that change. A fit to five activity indices gives 59.7565% to 112.6522% of the added amplitude over all phases. A calcium-only fit gives 98.9062% to 111.8543%. Exact response matrices agree with direct paired additions. Signals added only to later data give unit response and omit the effect on training. Formal night weights and exposure fits permit both loss and gain, but change their size. We preserve source tables, provenance, matrices, checks, figures, and offline reproduction code. The result measures the specified linear operation with fixed indices. It does not detect a planet or give a detection limit.

HD 26965 · NEID · radial velocity · stellar activity · signal injection · validation · linear regression · planet search