{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T08:48:30Z","timestamp":1791881310452,"version":"4.3.4"},"posted":{"date-parts":[[2026]]},"group-title":"SSRN","reference-count":10,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"abstract":"<jats:p>We adapt Two-Sided Projected Principal Component Analysis (TIP-PCA), a forecasting method originally developed for intraday volatility, to yield curve forecasting. The dynamic Nelson-Siegel (DNS) model, a standard benchmark, fixes the shape of its term-structure factors in advance and predicts each factor only from its own past. TIP-PCA removes both restrictions, estimating the maturity-side structure and the month-to-month dynamics nonparametrically, the latter driven by covariates built from past yield levels. This lets us build covariates from short, mid, and long maturity segments instead of the whole curve. Segmentation substantially improves accuracy over whole-curve covariates. With the rank set to three as in DNS, TIP-PCA attains lower RMSE than DNS, AR(1), and a random walk at the oneand three-month horizons. The gain over DNS is significant mainly at one month. At longer horizons, TIP-PCA does not improve on DNS, except marginally at twelve months with a validated rank. At rank three, the gains come mainly from the month-to-month dynamics rather than from estimating the maturity-side loadings, and run through more accurate slopefactor forecasts. The gains over DNS also widen in turmoil such as the 2007-2009 financial crisis and the recent high-inflation period, when DNS forecasts lag the yield curve while the flexible models track it more closely.<\/jats:p>","DOI":"10.2139\/ssrn.7553004","type":"posted-content","created":{"date-parts":[[2026,10,3]],"date-time":"2026-10-03T10:29:35Z","timestamp":1791023375000},"source":"Crossref","is-referenced-by-count":0,"title":["Dynamic Term Structure Forecasting with Two-Sided&amp;nbsp;Projected Principal Component Analysis"],"prefix":"10.2139","author":[{"given":"Seungwoo","family":"Choe","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2293-5080","authenticated-orcid":true,"given":"Bong-Gyu","family":"Jang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-0148-6692","authenticated-orcid":true,"given":"Jongho","family":"Park","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"ref1","doi-asserted-by":"crossref","first-page":"1203","DOI":"10.3982\/ECTA8968","article-title":"Eigenvalue ratio test for the number of factors","volume":"81","author":"S C Ahn","year":"2013","journal-title":"Econometrica"},{"key":"ref2","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1093\/jjfinec\/nbx002","article-title":"Forecasting bond yields with segmented term structure models","volume":"16","author":"C Almeida","year":"2018","journal-title":"Journal of Financial Econometrics"},{"key":"ref3","doi-asserted-by":"crossref","first-page":"258","DOI":"10.1111\/jtsa.12769","article-title":"Forecasting the yield curve: The role of additional and time-varying decay parameters, conditional heteroscedasticity, and macro-economic factors","volume":"46","author":"J F Caldeira","year":"2025","journal-title":"Journal of Time Series Analysis"},{"key":"ref4","doi-asserted-by":"crossref","first-page":"372","DOI":"10.1080\/07350015.2025.2537387","article-title":"Matrix-based prediction approach for intraday instantaneous volatility vector","volume":"44","author":"S H Choi","year":"2026","journal-title":"Journal of Business and Economic Statistics"},{"key":"ref5","doi-asserted-by":"crossref","first-page":"337","DOI":"10.1016\/j.jeconom.2005.03.005","article-title":"Forecasting the term structure of government bond yields","volume":"130","author":"F X Diebold","year":"2006","journal-title":"Journal of Econometrics"},{"key":"ref6","doi-asserted-by":"crossref","first-page":"193","DOI":"10.1002\/for.1258","article-title":"Forecasting the yield curve in a data-rich environment using the factor-augmented Nelson-Siegel model","volume":"32","author":"P Exterkate","year":"2013","journal-title":"Journal of Forecasting"},{"key":"ref7","doi-asserted-by":"crossref","first-page":"870","DOI":"10.1214\/12-AOAS551","article-title":"Functional dynamic factor models with application to yield curve forecasting","volume":"6","author":"S Hays","year":"2012","journal-title":"Annals of Applied Statistics"},{"key":"ref8","doi-asserted-by":"crossref","first-page":"65","DOI":"10.1002\/jae.2319","article-title":"Smooth dynamic factor analysis with application to the US term structure of interest rates","volume":"29","author":"B Jungbacker","year":"2014","journal-title":"Journal of Applied Econometrics"},{"key":"ref9","doi-asserted-by":"crossref","first-page":"20819","DOI":"10.1007\/s00521-023-08879-1","article-title":"Lagging problem in financial time series forecasting","volume":"35","author":"J Li","year":"2023","journal-title":"Neural Computing and Applications"},{"key":"ref10","doi-asserted-by":"crossref","first-page":"54","DOI":"10.3905\/jfi.1991.692347","article-title":"Common factors affecting bond returns","volume":"1","author":"R Litterman","year":"1991","journal-title":"Journal of Fixed Income"}],"container-title":[],"original-title":[],"deposited":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T08:04:59Z","timestamp":1791878699000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ssrn.com\/abstract=7553004"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":10,"URL":"https:\/\/doi.org\/10.2139\/ssrn.7553004","relation":{},"subject":[],"published":{"date-parts":[[2026]]},"subtype":"preprint"}}