সব বই

পেশাদার

1 Markets I: The Ecosystem and Exchange-Traded Marketsবাজার অনলাইন 2 Markets II: Rates, FX and Creditবাজার অনলাইন 3 Markets III: Commodities, Energy and Cryptoবাজার অনলাইন 4 Quantitative Methodsপদ্ধতি অনলাইন 5 Derivatives and Volatilityডেরিভেটিভস অনলাইন 6 Rates, Credit, XVA and Riskসুদহার, ঋণ ও ঝুঁকি অনলাইন 7 Research Craft: Predictors, Backtests, Measurement, Portfoliosগবেষণা অনলাইন 8 Strategies I: Equities and Futuresকৌশল অনলাইন 9 Strategies II: Volatility, Relative Value, Macro and the Bank Desksকৌশল অনলাইন 10 Microstructure and Executionএক্সিকিউশন অনলাইন 11 Market Making and High-Frequency Tradingমার্কেট মেকিং অনলাইন 12 Machine Learning for Marketsমেশিন লার্নিং অনলাইন 13 Low-Latency Softwareপ্রযুক্তি অনলাইন 14 Networks, Hardware and Trading Infrastructureপ্রযুক্তি অনলাইন 15 Research, Data and Risk Platformsপ্রযুক্তি অনলাইন 16 The Desk and the Firmপ্রতিষ্ঠান অনলাইন 17 The Industry: Firms, Roles and Careersক্যারিয়ার অনলাইন 18 Interviews I: Maths, Probability and Speedক্যারিয়ার অনলাইন 19 Interviews II: Markets, Trading Games and the Final Roundক্যারিয়ার অনলাইন 20 Interviews III: Codeক্যারিয়ার অনলাইন
অ্যাপ পরিচিতি Coach সাইন ইন পড়া শুরু করো

বইগুলো এখনো বাংলা ভাষায় পাওয়া যায় না — ইংরেজি সংস্করণ দেখানো হচ্ছে।

Mathematics · পরিভাষা

Bounds, sup and inf কী?

অন্য নাম: upper bound

Definition 10.1 University Mathematics — Year 1 · অধ্যায় 10 — Real Numbers

Let A⊆RA \subseteq \R be nonempty. A real MM is an upper bound of AA when a≤Ma \leq M for all a∈Aa \in A; AA is bounded above when it has an upper bound (similarly below, with lower bounds; bounded means both). A maximum of AA is an upper bound belonging to AA.

The supremum sup⁡A\sup A is the least upper bound of AA, when it exists; the infimum inf⁡A\inf A is the greatest lower bound.

The set A = \1 - 1n : n ∈ ℕ*\ on the number line: its points pile up toward 1 without reaching it. Every number ≥ 1 is an upper bound (the ray), and nothing smaller is, because an element of A enters each interval (1 - , 1): the two clauses of  in one picture. The supremum is the left endpoint of the ray of upper bounds — and the completeness axiom is precisely the guarantee that this ray always has a left endpoint.
The set A={1−1n:n∈N∗}A = \{1 - \frac1n : n \in \N^*\} on the number line: its points pile up toward 11 without reaching it. Every number ≥1\geq 1 is an upper bound (the ray), and nothing smaller is, because an element of AA enters each interval (1−ε, 1)\intoo{1 - \varepsilon}{1}: the two clauses of Proposition 10.4 in one picture. The supremum is the left endpoint of the ray of upper bounds — and the completeness axiom is precisely the guarantee that this ray always has a left endpoint.

উদাহরণ

Example 10.5

sup⁡(0, 1)=1\sup \intoo{0}{1} = 1, not attained (no maximum); sup⁡[0, 1]=1=max⁡\sup \intcc{0}{1} = 1 = \max. For A={1−1n:n∈N∗}A = \{1 - \frac 1n : n \in \N^*\}: sup⁡A=1\sup A = 1, not attained; inf⁡A=min⁡A=0\inf A = \min A = 0. A maximum, when it exists, is the supremum; the whole point of sup⁡\sup is to have a substitute when the maximum does not exist.

Example 10.6 (Computing suprema in practice)

Two full workouts of Proposition 10.4.

The set A={x+1x:x>0}A = \{x + \frac1x : x > 0\}. For every x>0x > 0, x+1x−2=( x−1/x )21≥0x + \frac1x - 2 = \frac{(\,\sqrt x - 1/\sqrt x\,)^2}{1} \geq 0, so 22 is a lower bound; and 2=1+11∈A2 = 1 + \frac11 \in A: therefore inf⁡A=min⁡A=2\inf A = \min A = 2, attained at x=1x = 1. Above, AA is unbounded (x+1x>xx + \frac1x > x can exceed any MM by Theorem 10.10): sup⁡A\sup A does not exist in R\R (it is +∞+\infty in R‾\overline\R).

The set B={mm+n:m,n∈N∗}B = \bigl\{\frac{m}{m + n} : m, n \in \N^*\bigr\}. Every element lies in (0, 1)\intoo{0}{1}, so 00 and 11 are bounds. Neither is attained: mm+n=1\frac{m}{m+n} = 1 would force n=0n = 0. For the supremum, freeze n=1n = 1 and let mm grow: mm+1=1−1m+1>1−ε\frac{m}{m+1} = 1 - \frac{1}{m+1} > 1 - \varepsilon as soon as m+1>1εm + 1 > \frac1\varepsilon (Archimedes): sup⁡B=1\sup B = 1. Symmetrically (m=1m = 1, nn large), inf⁡B=0\inf B = 0. The closing insight: to pin a supremum, one well-chosen one-parameter path inside the set suffices — here the path n=1n = 1 — and the ε\varepsilon-characterization asks for nothing more.

Example 10.7 (The infimum mirror)

The infimum has its own ε\varepsilon-characterization, obtained from Proposition 10.4 through inf⁡A=−sup⁡(−A)\inf A = -\sup(-A): i=inf⁡Ai = \inf A iff ii bounds AA below and, for every ε>0\varepsilon > 0, some a∈Aa \in A has a<i+εa < i + \varepsilon. A workout with both bounds at once: let

A={(−1)n+1n:n∈N∗}={0, 32, −23, 54, −45, … }.A = \Bigl\{(-1)^n + \frac1n : n \in \N^*\Bigr\} = \Bigl\{0,\ \tfrac32,\ -\tfrac23,\ \tfrac54,\ -\tfrac45,\ \dots\Bigr\} .

Even indices give 1+1n≤321 + \frac1n \leq \frac32, with equality at n=2n = 2: since also the odd-index values are ≤0<32\leq 0 < \frac32, we get sup⁡A=max⁡A=32\sup A = \max A = \frac32. Odd indices give −1+1n>−1-1 + \frac1n > -1, decreasing toward −1-1: every element of AA is >−1> -1, and −1+ε-1 + \varepsilon is beaten by −1+1n-1 + \frac1n for odd n>1εn > \frac1\varepsilon: inf⁡A=−1\inf A = -1, not attained. One set, all four behaviors on display: a supremum that is a maximum, an infimum that is not a minimum.

প্রসঙ্গসহ পড়ো →