Author
When inflation is quiet, rates still follow growth. Money supply, consumer confidence, and unemployment duration read the economy's speed, strongest on the long end, and only when they agree.
Lenders demand a real return, so short rates track inflation. The Real-Rate Ratio turns that into one number, and a negative reading is the loaded spring that preceded the 1993 bond bear.
A vetted seasonal is still a weak 56/44 bias that dies to costs if traded alone. Its real job is to gate or tilt a system you trust, stacking with intermarket filters as one input among several.
Five days, two directions, endless mining. The S&P Monday bias decayed, coffee's Thursday is a guess, but silver's Thursday survives because it is really an economic-strength signal in disguise.
Most seasonals are curve-fits with good marketing. Four tests separate a real calendar edge from a coincidence: beat the drift, survive dropping the best year, beat the coin, and name the cause.
Three ways to compute a seasonal, raw, detrended, and standardized, give three different curves from the same data. The method decides what the number means, and one of them quietly leaks the future.
Seasonals come from three different machines: fixed dates, floating events, and human habit. Name which engine you are trading and you know how it will break.
Predicting gross return secretly bundles a market-timing bet you're bad at. Strip out beta times the market and other exposures, forecast only the idiosyncratic residual, and the job gets honest.
MSE on variance is the wrong ruler for vol forecasts. QLike is optimal without knowing the distribution and punishes underestimating vol more than overestimating, matching the maker's adverse-selection cost.
A factor is just an alpha that explains so much variance you keep re-finding it. Subtract it from returns, then dot your positions with what's left: flat means old exposure, sloped means real edge.
Every indicator is a filter: a machine that reshapes price. Learn signal, frequency, lag, and the four filter jobs once, and the whole cycles literature stops being a wall.
Every FX pair blends two currencies. Here is how to un-mix them with one matrix and ridge least squares, why it reconstructs every pair at 0.999, and why that still is not an edge.