?⃝ Exam # 1 Continuous Random Variables Independent if PLAIBT-PCAHPCBM-t-plblpcn-nbt-pcn.mn Two : g.gg# The probabability oÑ* The , can't pmflpdf - . happen at the sometime (NEVER INDEPENDENT ) mass function pofa discrete random variable Flik pcat-PLK-alfor-ascac.es bound to /flxldx derivative of dist function . dist function . integral pmflpdf / Ill " median 1m fly)dx= → )={ FIX - too -11×1=1 interval , make sure to Of the intervals -_ ✗ = - f- "× 42 - PlX≤ a) -5lb ) = - Flat Finding cdfofapmflpdfata specific Pla≤X≤ b) =/abtcxldx Quantile :Flqp1=PlX≤ qp / =p • Pla < ✗ ≤ b) =P( ✗ ≤b) Note : or continuous probability density function 1-1×1=1--4×1 . . . Disjoint no overlap Discrete : countable Plxtakplx≥ a) events : MAMAN .nAm1= MADMAN .PlAm) or more ✗ add the full integral below it Go lower on each . interval . ✗ ≤ -3 0 44%4-4--21=411--1%4 dx -3*-2 % " " ≤3 ' Yadx -z≤x≤z %%dx+fj%dx2≤x≤ 3 I I ×≥3 #2 EXAM flxldx 42 FIXKYZ -_ Median Discrete 1) Bernoulli distribution Blrlp) 2) PlX=l/ =p where o≤p≤ I PIX-01=1 p E- [X] =p Varlxtpltp) Binomial distribution Binln,p) , where 0≤p≤ I - PlX=k)=( 1) pklt-pln-kfork-ql.int/T=npandVarlX)=npll-p) 3) Glow, Where 04pct Geum where 01Pa 4) Poisson distribution Poisltll where µ 0 PLK-kt-%Fe-ntork-o.li plk-kt.pl/-p1K-'torlh- lR.- ElXI=YpandVarlx1- 1-p1/p2 > , FIX]=MVarN=M , . continuous 5) Exponential Dist Expat , where d> 0 6) Normal dist NIMOY where ounces and E[X]=M VARIXKOZ use 2-table . "× flH=Xl _ FIX)=l - e- 7) Paretodist Park), where . ¥+1 FIX)=tX fork≥ / - "× E[X]=y× vary)=yµ for ×≥o - . NO ✗ for ]=✗/L✗- 1) E- [× ×≥ / VARIX)=✗ /((d- 1)4×-2)) fort> 2 and for a > land 8) Uniform distribution Ulaid , where forocxa flH=É as for 0 < ✗≤ I for ✗ ≤ ✗ ≤b 1=411--5-8 acb for a≤x≤ b E[X]=(a+bk Varlxtfb Expectation ? variance discrete :E[glXD¥glailMX=ai) ANTINOUS :-[[×] =/ - f.[ (✗ +412]=E[✗2) • Xflxldx EEY as + ]=[ yflyldy - E[Y2]t2E[XY] COVLXY )=E[XY ] (E[X]E[Y]) - play)= COVLXY) varcxivarcy, E[X+y7=E[X]tE[y] as E[X7=f%fWdX - WANT ]=E[T2] (ECT])2 - covariance correlation coefficient varlvtwt-varl) vltvarlwlvarlrwt-rzvarlwvarl-YT-varlxtyln.ro '2=var(X A- varies Jensen's independent binomial 2- HEAD npq Inequality g. LEAD 1=[91×1] concave gleam ≥E[gH)] Dependent if : planB) =/ PLAIPLB) )≤ convex the law of large numbers : Chebyshev's inequality : ]=u&VartXn=% E[In PAY EEN ≥ a) ≤ atvarly) - PHY Ely] / ≤ at ≥ - 2- scores : Zn=Tn④-U) o m .­ metrical 0>0 Zn=In-E[In] Varun 1- v%# 2n= ✗ it . . . .tl/n-npl Orn as ECXY ]=y)dydx a a. - a) 412 suppose that is known that the number of items produced in a factory during a week is a random variable with mean so a) What can be said about the probability that this week's production will exceed 75? b) If the variance of a week's production is known to equal 25, then what can be said about the probability that this week's production will be between 40 & 60? . Let X be the number of items that will a) By Markov's inequality PEX>75} ≤ Ef = Hence {1×-501<101 } and so the - probability produced in a week : 5¥ } = b) By Chlbshev's inequality PE 1×-501≥ to} ≤ , ≥ I be = 4- Ya 314 that this week's = 40 and 60 is at least 75 • . production will be between