Conditional Density Estimation and Density Forecast With Applications
Bibliographic record
Abstract
In this thesis, several univariate and multivariate methods of building conditional density estimates and density forecasts are considered.For each method, two approaches (one-step and two-step) of finding the optimal bandwidth windows are used.To check which method is more accurate, numerous simulation studies were done on different univariate linear models such as AR(1) and multivariate linear models such as AR(2).Nonlinear models were also studied such as AR(1) with ARCH errors.These ideas were later applied to real world data where the true densities are unknown and checked to see how good they forecast the next point.Stock market data was used where different methods of the multivariate one-step approach were applied.The stocks that were studied were the NYSE and TSX.The dependence of one stock market on the other was also studied and how much of a factor it plays in forecasting the next price.Weather data was also studied.The three weather variables studied were the daily averages of pressure,wind speed and temperature for the Ottawa region.Multiple forecast densities were built and compared to see which method has the highest accuracy.It was found that when dealing with non-linear data, the one-step method is more efficient in estimating the true conditional density.The two-step approach is better when trying to estimate a univariate conditional density.When trying to estimate multivariate conditional density the further one goes away from stationarity, the better that the 2-step approach is.Real world data indicates that multivariate methodology was better at predicting future events.3.4 Multivariate Response . . . . . . . . . .
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.033 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.002 | 0.004 |
| Insufficient payload (model declined to judge) | 0.006 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".