{"entities":{"Q6896436":{"pageid":20875129,"ns":120,"title":"Item:Q6896436","lastrevid":84185933,"modified":"2026-05-13T07:56:28Z","type":"item","id":"Q6896436","labels":{"en":{"language":"en","value":"A finiteness condition for complex continued fraction algorithms"}},"descriptions":{"en":{"language":"en","value":"scientific article; zbMATH DE number 8125092"}},"aliases":{},"claims":{"P31":[{"mainsnak":{"snaktype":"value","property":"P31","hash":"fd5912e4dab4b881a8eb0eb27e7893fef55176ad","datavalue":{"value":{"entity-type":"item","numeric-id":56887,"id":"Q56887"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q6896436$C22A1D5F-2408-4888-9E9F-119070908FC6","rank":"normal"}],"P159":[{"mainsnak":{"snaktype":"value","property":"P159","hash":"5c73e631e9a7f744c809c98f67503cd962fc9a72","datavalue":{"value":{"text":"A finiteness condition for complex continued fraction algorithms","language":"en"},"type":"monolingualtext"},"datatype":"monolingualtext"},"type":"statement","id":"Q6896436$03B0AA8E-D288-4DDF-95B4-7F8F599D0E17","rank":"normal"}],"P27":[{"mainsnak":{"snaktype":"value","property":"P27","hash":"bdcfbd949798af87ba79d78ac608cfe3f9196a2c","datavalue":{"value":"10.1090/PROC/17380","type":"string"},"datatype":"external-id"},"type":"statement","id":"Q6896436$D49C5419-B6CF-4169-9AD0-CDCB26806821","rank":"normal"}],"P16":[{"mainsnak":{"snaktype":"value","property":"P16","hash":"0df9d5a70ab2d7c962ceb6330b5a0cb35bf5693c","datavalue":{"value":{"entity-type":"item","numeric-id":405733,"id":"Q405733"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q6896436$D9536470-E841-4735-8DBA-C31F38B3D089","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P16","hash":"e06da32ca064b8643fffd920efaf38076b6eb30b","datavalue":{"value":{"entity-type":"item","numeric-id":1661135,"id":"Q1661135"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q6896436$A4F2B65C-08DE-45EB-AA33-E5DB9A064837","rank":"normal"},{"mainsnak":{"snaktype":"value","property":"P16","hash":"ddb204124299e58810c9e5c8baf6911cdd793db2","datavalue":{"value":{"entity-type":"item","numeric-id":331133,"id":"Q331133"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q6896436$48D35CE5-D008-4C38-840D-B41ECFD01ED0","rank":"normal"}],"P200":[{"mainsnak":{"snaktype":"value","property":"P200","hash":"e7b55aa1a8dafd08ef90d95361f82f1155bd5da7","datavalue":{"value":{"entity-type":"item","numeric-id":115236,"id":"Q115236"},"type":"wikibase-entityid"},"datatype":"wikibase-item"},"type":"statement","id":"Q6896436$EDE2E7B6-D4C4-4A1E-8388-0F8AA2451CB0","rank":"normal"}],"P28":[{"mainsnak":{"snaktype":"value","property":"P28","hash":"4f974a567496d997696a7f7f0ff44cd54be3aa7c","datavalue":{"value":{"time":"+2025-11-21T00:00:00Z","timezone":0,"before":0,"after":0,"precision":11,"calendarmodel":"http://www.wikidata.org/entity/Q1985727"},"type":"time"},"datatype":"time"},"type":"statement","id":"Q6896436$D39C8906-53B8-4A58-855A-7D64EEBE7361","rank":"normal"}],"P1448":[{"mainsnak":{"snaktype":"value","property":"P1448","hash":"6652a683873848f06be8485864bde4cbd44b4521","datavalue":{"value":"The paper is concerned with \\textit{continued fraction algorithms} that allow a representation of real or complex numbers by digits (the \\textit{partial quotients}) \\(a_1,a_2,a_3,\\ldots\\) taken from an infinite set \\N\\[\\Nz=\\cfrac{1}{a_1+\\cfrac{1}{a_2+\\cdots}}. \\N\\]\\NThe authors give a historical overview from Euclid's algorithm (iteration of the Gauss map) to other types, pointing out connections with Lagrange's best approximation,\\textit{nearest integer continued fractions}, several types Hurwitz algorithms etc.\\N\\NThe layout of the paper is as follows:\\N\\N\\S1 Introduction (\\(3\\frac{1}{2}\\) pages).\\N\\N\\S2 The Hurwitz algorithm (\\(1\\frac{1}{2}\\) pages).\\N\\NHere papers by \\textit{A. Hurwitz} [Acta Math. 11, 187--200 (1888; JFM 20.0201.01)] and \\textit{J. Hurwitz} [Acta Math. 25, 231--290 (1902; JFM 33.0221.04)] are referred to; it is a generalization of the nearest integer algorithm for numbers in the complex plane.\\N\\NAs usual the set of \\textit{Gaussian integers} is given by\\N\\[\\N\\mathbb{Z}[i]=\\{a+bi\\,:\\,a,b\\in\\mathbb{Z}\\},\\ i=\\sqrt{-1};\\ U=\\{z\\in\\mathbb{C}\\,:\\,-\\frac{1}{2}\\leq\\text{Re}\\,z,\\,\\text{Im}\\,z<\\frac{1}{2}\\}.\\N\\]\\N(\\(\\text{Re}z\\) and \\(\\text{Im}z\\) are the real and imaginary parts of \\(z\\) respectively)\\N\\NThe Hurwitz algorithm is defined through the iterates of the Hurwitz map\\N\\[\\NT\\,:\\, U\\setminus \\{0\\}\\rightarrow U;\\ z\\mapsto \\frac{1}{z}-\\lfloor \\frac{1}{z}\\rfloor_U\\tag{*}\\N\\]\\Nwhere the floor function \\(\\lfloor . \\rfloor_{1/2}\\) is replaced by \\(\\lfloor\\cdot\\rfloor_U : \\mathbb{C}\\rightarrow U\\) with \\(z\\rightarrow z-w\\), \\(w\\) the unique Gaussian integer satrisfying \\(z-w\\in U\\).\\N\\NFor a given \\(n\\)-tuple \\(b_1,b_2,\\ldots,b_n\\in\\mathbb{Z}[i],\\ n\\geq 2\\), the \\textit{cylinder}-sets \\(\\Delta(b_1,\\ldots,b_n)=T^{-1}_{b_1}(\\Delta(b_2,\\ldots,b_n))\\) play an important role (here \\(b\\in\\mathbb{Z}[i],\\ \\Delta(b)=\\{z\\in U\\, :\\, \\lfloor\\frac{1}{L}\\rfloor_U=b\\}\\) and \\(T_b\\) is the restriction of \\(T\\) to \\(\\Delta(b)\\)).\\N\\NThus \\(\\Delta(b_1,\\ldots,b_n)\\) is the set of points \\(z\\in\\mathbb{Z}[i]\\) that have the first digits \\(a_i=b_i\\ (1\\leq i\\leq n)\\) in their Hurwitz expansion.\\N\\N\\S3 A generalization: the \\(\\alpha\\)-Hurwitz algorithm (\\(2\\frac{1}{2}\\) pages).\\N\\NHere the main object of the paper appears: a perturbation of the Hurwitz map. For \\(\\mathbf{\\alpha}=(\\alpha_1,\\alpha_2) \\in \\mathbb{R}^2\\), \\(T\\) in formula \\((\\ast)\\) is made to depend on the \\textit{shifted} domain \\N\\[\\NU_{\\mathbf{\\alpha}}=\\{z\\in\\mathbb{C}\\, :\\, \\alpha_1-1\\leq \\text{Re} z <\\alpha_1,\\ \\alpha_2-1\\leq \\text{Im} z <\\alpha_2\\},\\]\\Nand \\(\\lfloor\\cdot\\rfloor_{\\mathbf{\\alpha}}\\) is the function \\N\\[\\N\\lfloor\\cdot\\rfloor_{\\mathbf{\\alpha}} : \\mathbf{C}\\rightarrow U_{\\mathbf{\\alpha}},\\ z\\mapsto z-w,\\N\\]\\Nwhere \\(w\\in\\mathbb{Z}[i]\\) is the unique Gaussian integer with \\(z-w\\in U_{\\mathbf{\\alpha}}\\).\\N\\NThe \\(\\mathbf{\\alpha}\\)-Hurwitz map is given by \\N\\[\\NT_{\\mathbf{\\alpha}} : U_{\\mathbf{\\alpha}}\\setminus \\{0\\} \\rightarrow U_{\\mathbf{\\alpha}},\\ z\\mapsto \\frac{1}{z}-\\lfloor \\frac{1}{z}\\rfloor_{\\mathbf{\\alpha}}.\\N\\]\\NThe main result is now\\N\\NTheorem 3.1. Let \\(p,q,r,s\\in\\mathbb{N}\\) be such that \\((\\frac{p}{q},\\frac{r}{s})\\in\\mathcal{D}\\). For \\(\\mathbf{\\alpha}= (\\frac{p}{q},\\frac{r}{s})\\) the \\(\\mathbf{\\alpha}\\)-Hurwitz algorithm satisfies the finite range condition. Equivalently, the map \\(T_{\\mathbf{\\alpha}}\\) admits a partition with finitely many atoms of \\(U_{\\mathbf{\\alpha}}\\).\\N\\N\\S4 Proof of the main result (\\(3\\frac{1}{2}\\) pages).\\NProof of Theorem 3.1.\\N\\N\\S5 Perspectives (\\(1\\) page).\\NTouches upon different choices of the \\(\\alpha_1,\\alpha_2\\), the optimal constant \\(L\\) for which the convergents of the \\(\\mathbf{\\alpha}\\)-Hurwitz algorithm satisfy \\(|z-p_n/q_n|