Coursehero >>
New York >>
Rochester >>
CSC 258 Course Hero has millions of student submitted documents similar to the one below including study guides, homework solutions, papers, and exam answer keys.
eServer: IBM (Discovery) Power4 System Micro-architecture A Summary Paper goes into a lot of detail about the core. Branch Prediction is paramount to long pipelines. Aggressive instruction fetching and pre-caching. Groups instructions into groups to ease logic of branch/cache misses. Also, marketing about 32/64 bit support, iSeries, pSeries, etc. not real interesting for us. 252, 254 stuff. By Thomas Thomas Multi-processor Organization 2 processing cores per chip. 4 chips per module 1, 2, 3, or 4 modules per SMP. That s up to 32 hardware contexts running concurrently Each chip has it s own cache and cache controllers for system balance. Processor Communication Logically there are four, 16-byte buses for on module communication. Implemented with 6 buses: 3 on and 3 off There are two, 8-byte buses for inter-module communication. 1 on and 1 off Other buses for L3, IO devices, and system service processor. (GX bus) All processors use their own bus, and snoop everyone else s. Chip Organization 1 Out of Order Load/Store Unit and Cache Coherence Decode, Crack and Group Formation: As instructions are executed out of order, it is necessary to remember the program order of all instructions in flight. In order to minimize the logic necessary to track a large number of in flight instructions, groups of instructions are formed. The individual groups are tracked through the system. That is, the state of the machine is preserved at group boundaries, not at an instruction boundary within a group. Any exception causes the machine to be restored to the state of the oldest group prior to the exception. Continued Even with instructions in groups there is a lot of book keeping, with mundane instructions, others need special attention. Load, Hit, Store: Load executing before older store finishes writing. (forwarded data) Store, Hit, Load: Load executing before we realize there is an older store. (stale data) Load, Hit, Load: Old/new load consistency. Memory Hierarchy and coherency The L2 Cache The coherency point for POWER4 is the L2 cache. All data in the L1 data cache is also in the L2 cache. The coherency processor Controls the return of data from the L2 (hit) or from the fabric controller (miss) to the requesting core via the CIU; Updates the L2 directory as needed; Issues fabric commands for L2 misses on fetch requests and for stores that do not hit in the L2 in the M, Me or Mu state (described below); Controls writing into the L2 when either due reloading to fetch misses in the L2 or stores from the processors; and, Initiates back invalidates to a processor via the CIU resulting from a store from one core that hits a cache line marked as resident in the other processor s L1 data cache. Coherency States (different from Sun s MESI) I (invalid state): Invalid. SL (shared state, can be source to local requesters): S (shared state): Somebody else will share M (modified state): Exclusively owned, but can be sourced to others. Me (exclusive state): Not considered modified, but I have requested exclusivity. Mu (unsolicited modified state): Modified, and still exclusive. T (tagged state): Was modified and sourced to another L2. 2 The L3 Cache. The L3 is designed to be used as a standalone 32 MB L3 cache, or to be combined with other L3s on the same processor module in pairs or groups of four to create a larger, address interleaved L3 cache of 64 MB or 128 MB. Combining L3s into groups not only increases the L3 cache size, but also scales the available L3 bandwidth L3 Continued I (invalid state): The data is invalid. S (shared state): The data is valid. In this state, the L3 can only source data to L2s that it is caching data for. T (tagged state): The data is valid. The data is modified relative to the copy stored in memory. The data may be shared in other L2 or L3 caches. Trem (remote tagged state): This is the same as the T state, but the data was sourced from memory attached to another chip. O (prefetch data state): The data in the L3 is identical to the data in memory. The data was sourced from memory attached to this L3. The status of the data in other L2 or L3 caches is unknown. The Module Multiple Modules Multiple Module Interconnect: Figure 9 shows the interconnection of multiple 4-chip MCMs to form larger SMPs. From 1 to 4 MCMs can be interconnected. When interconnecting multiple MCMs, the intermodule buses act as repeaters moving requests and responses from one module to another module in a ring topology. As with the single MCM configuration, each chip always sends requests/commands and data on its own bus but snoops all buses. Multiple Modules, continued. Power5 184 million transistors to 276 million transistors Increased cache size and bandwidth L3 now on processor side of fabric controller And off the inter-chip bus Simultaneous multi-threading per core. 2 threads. Up to 8 modules, or 128 active threads. 4 books * 2 MCMs * 4 Processors * 2 cores * 2 threads 3
Find millions of documents here - Study Guides, Homework Solutions, Papers, Exam Answer Keys and more.
Course Hero has millions of course related materials that will enable you to learn better, faster and get an A in all your courses.
Below is a small sample set of documents:
258_TreadMarks_jonathan.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: TreadMarks: Shared Memory Computing on Networks of Workstations Authors: Cristiana Amza, Alan L. Cox, Sandhya Dwarkadas, Pete Keleher,Honghui Lu, Ramakrishnan Rajamony, Weimin Yu, Willy Zwaenepoel Presented by:Jonathan Williams Why a Distributed Sha...
Carpenter_HPC.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: HighPerformanceComputing: BlueGene&Cray AaronCarpenter March28th,2007 CSC458 3/28/07 CarpenterHPC Outline HPC BlueGene/L Processingmodule Interconnect Communicationprimitives Cray CellProcessor 3/28/07 CarpenterHPC HPCTop500List X1 BlueG...
proj3.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: XXt@ !hs#8t XXt@ X@t k p x fk i f f G #EXu6s t@G`XEyf kpXie`Ys @tqG#`P#W`Yw58VX`WX`}&p) n t U I Y P x d t x E P WG v W xYYI i Efa x Wi EP WId Ef ehhXX`h! k UG x fi YGYY d G U f PYI U ...
09_pred_logic.pdf
Path: Rochester >> CSC >> 173 Fall, 2009
Description: Predicate Logic Limitation of propositional logic no way to talk about properties that apply to categories of objects, or relationships between those properties Predicate logic mathematical model for reasoning about predicates: functions that map v...
saurabh_capriccio.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: Capriccio:ScalableThreadsfor InternetServices RobvonBehren,JeremyCondit,FengZhou, GeorgeC.Necula,EricBrewer Presentedby SaurabhDeshpande CS458(03/07/07) Motivation Tocreatescalablehighlyconcurrentservers Developathreadbasedsystemasagainstaneventba...
vinod_SEDA.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: SEDA Staged Event-Driven Architecture Problem/Motivations: Internet applications catering to bursty, massively concurrent demands Responsive Robust Fault tolerance Concurrent requests translate to even higher I/O and N/w requests Example: 1.2 billio...
sam_group_comm.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Improving Scalability for Reliable Group Communication Sam Shaojun Zhao Outlook Group Communication Scalability Issue Previous Approaches Gossip-Based Protocol Security Issues Tools and Applications Conclusion Group Communication Why grou...
bhardwaj_Consistency_Models.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: Outline Shared Memory Consistency Models Authors: Sarita V. Adve and Kourosh Gharachorloo - Nitin Bhardwaj What is Memory Consistency Sequential Consistency Optimizations to SC Relaxed Memory Consistency Processor Consistency Week Consisten...
wilbur_mcs.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: Algorithms for Scalable Synchronization on Shared Memory Multiprocessors John Mellor-Crummey and Michael L. Scott Greg Wilbur February 2, 2009 Overview Locks test and set lock Ticket Lock Array Based Queueing Locks MCS Lock Barriers Centralized Barr...
lecture10.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Concurrency A process or thread is a potentially active execution context Classic von Neumann model single thread of control, single active execution context Concurrency two or more execution contexts may be active at the same time Why concurr...
lecture8.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Logic Programming Imperative programming models: compute via iteration and side effects Functional programming model: compute via recursion and substitution of parameters into functions Logic programming model: compute via resolution of logical s...
mh_thread_implementation.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Thread Analysis and Implementation CSC 258 Parallel & Distributed Systems January 31st, 2005 Outline Thread Management for Shared Memory Concurrency refresher Alternatives to spinspinlocks Spin-lock Spinmanagement alternatives Dynamic Thread Implem...
05_grammars.pdf
Path: Rochester >> CSC >> 173 Fall, 2009
Description: Context-Free Grammars Adds recursion/allows non-terminals to be expressed in terms of themselves Can be used to count/impart structure e.g., nested parentheses Notation grammar G(S,N,T,P) S is the start symbol N is a set of non-terminal symb...
lecture2.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Summary What is a programming language What makes a language successful Why study programming languages Programming language views Types of languages Phases of compilation Scanners and Parsers Types of languages Imperative Von Neumann (Fort...
lecture6.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Intermediate Representations Structural (graphically oriented) E.g. abstract syntax tree, directed acyclic graph Ordering Control Flow Linear E.g., stack machine/one-address code, 3address code Hybrids E.g., control flow graph Basic Paradi...
hai_checkpointing.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Rollback-Recovery Protocols in Message-Passing Systems Hai Nguyen Approaches to reliability - Transactions: data-oriented applications - Group communications: abstraction of ideal communication system - Rollback recovery: long-running applications ...
04_fa_and_re.pdf
Path: Rochester >> CSC >> 173 Fall, 2009
Description: Patterns, Automata, and Regular Expressions Finite Automata formal or abstract machine to recognize patterns Regular expressions formal notation to describe/generate patterns Finite Automata A finite collection of states An alphabet A set of...
08_prop_logic.pdf
Path: Rochester >> CSC >> 173 Fall, 2009
Description: Propositional Logic CSC 173 Propositional logic mathematical model (or algebra) for reasoning about the truth of logical expressions (propositions) Logical expressions propositional variables or logical expressions connected with logical operato...
time_global_state.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Distributed Systems A collection of independent (autonomous) computers that appear as a single coherent system E.g., world wide web, distributed file systems Time and Global States Properties Transparency Scalability Availability Reliabilit...
lecture5.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Types Denotational view Set of values (e.g., enumerated type) Type Systems Constructive view One of a small collection of built-in types or a composite type (record, array, set, ) Abstraction-based view An interface consisting of a set of op...
cs254_overview.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Types of languages Imperative CS254: Programming Language Design and Implementation Overview Von Neumann (Fortran, Pascal, BASIC, C, ) Object-oriented (C+/Java, Smalltalk, ) Declarative Functional (Scheme, Lisp, ML, ) Logic, constraint-based...
assign-4.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: a D D S! 0 C % ! F C 6 \' # U \' R C f a % # \' D w D # % S R C S % X # 6 \' 6 C f p $#\"4$D\"Hs$!r!hfphHj\'hw #1rH@ErC$!`@hHi\"r4rRk1r$R1r#\"E\"vrHj\'\"1R r#xi@H1$#\"H C S C % S \' F p n ! D R U C D p D # ! C % ! ...
assign-3.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: h ) 8 { # ) qD)( d) u(w 3 je ({ n{ % u%C T%\"# ) @ C D $ 6e i # 3 e w 1h @l i 3 ho 68 $ \'\"$) f\'1)(! 3 )9j8 3 \"#d# ...
assign-1.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: 8 m # ) 8 k e # 5 k 8 C) ) C D ) e k $ 8 r3%yn3%3)%G3%y%\"$\"#\'%RiGG%rB%gG) 1u3i ) C) ! $ j e # m # C C C Dq $ 5 8 2 8 ...
gulcin_mdc.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: Image and Video Compression CS457 Seminar Gulcin Caner November 25, 2001 Outline Introduction Lossless compression JPEG (Still image compression standard) Video compression standards Multiple description coding (MDC) What makes compression possible...
boutell_rtp.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: Fast-forward to the Year 2021 Real-time Transport Protocol Matt Boutell CS457: Computer Networks November 15, 2001 Director of Development for MME, Inc. You Common Service Video Conferencing Streaming Audio Movies ? Two Goals of RTPs Common Serv...
oztan_CScong.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: Outline Congestion Control for Streaming Media over Internet Oztan Harmanci . . . . Streaming Data Predictable jitter and delay Certain bandwidth Multicast/Unicast Efficient encoding/limited complexity Real time/stored Video/audio: precisi...
proj1.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: !|o o1\'y|1Hvh w|yyoh1\'|wFx6o|yx E~6HU%h)gjED)8\'!x)e6zED6H6www6(!6)ED@)6C s C$C i 7 \" 2 DC $ ~ ~ ~ u $u(( i s g l $ g s \" j D$u& s i ...
assign-2.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: Qc$j9jeQsjfWjf\'sjec!6qscj q9fh \'4qsjjgsWjfWshjhjqG$qsAeq sA6sQetjejxjccje e 5 2 2 ( 2( ( 2 ...
proj3.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: n 6 B $ i $ i i di 2 i $ B iC )jp\'5!%k5G)U!\'kp\'G)\'kSk6D7m5G)kdD#GB d \" 6 $ 6 i @ 4 ( 2$ ( \" 6 6 B i i kCU5...
proj2.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: 4( 2$ $ $ } B$ B g ( g 6%eh\'ei F F H ~ B $ } ( ~ gC \" 6 C m $ ( ( \" 6 $ ( \" 6 B ( )5U5YsnG~Y)%D%59n7\'5s% )\'eGs6p\'5%5x)5xYh CB } B & f...
wanli_firewalls.pdf
Path: Rochester >> CSC >> 257 Fall, 2009
Description: Curious hackers Vandals Industrial spies Accidental data disclosure Solution? A computer or computer software that prevents unauthorized access to private data (as on a companys local area network or intranet) by outside computer users (as...
zhao_treadmarks.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: Treadmarks Shared Memory Computing on Networks of Workstations Hongzhou Zhao 2007/3/21 Treadmarks: Shared Memory Computing Distrbuted shared Memory System - Consist of N networked workstations, each with its own memory, connected by a network - A g...
lecture4.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: Questions not answerable by a CFG Semantics: Context-Sensitive Analysis Is x a scalar, an array, or a function? Is x declared before it is used? Are any names declared but not used? Which declaration of x does this reference? Is an expre...
lecture3.pdf
Path: Rochester >> CSC >> 254 Fall, 2009
Description: LR Parsers SLR(1) LR(0) items (construct DFA for recognizing viable prefix with no lookahead), use FOLLOW information to More guide reductions general LALR(1) merge states of LR(1) with same core (LR(0) items) LR(1) store lookahead information ...
lecture-4.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Where we are Parallelism, dependences, synchronization Patterns of parallelism data parallelism task parallelism Programming Languages Concurrent languages e.g., Occam, SR, Java, Ada Compiler-supported extensions e.g., HPF Library packages ...
lecture-3.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Basics of Parallelization Dependence analysis Synchronization Events Mutual exclusion Steps in the Parallelization Decomposition into tasks Expose concurrency Assignment to processes Balancing load and maximizing locality Orchestration N...
veronda_ssd_logfs.pdf
Path: Rochester >> CSC >> 256 Fall, 2009
Description: File Systems of Solid-State Drives A Brief Teaching by Mark Veronda Tuesday, November 4, 2008 Tuesday, November 4, 2008 1 Outline Background Solid-State Drives (SSD) History and Motivations Properties Implications Log-Structured (LS) File S...
shared_memory.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: Processes or Threads A process or thread is a potentially-active execution context Processes/threads can come from Multiple CPUs Kernel-level multiplexing of single physical CPU (kernel-level threads or processes) Language or library-level multi...
2005-10_midterm.pdf
Path: Rochester >> CSC >> 173 Fall, 2009
Description: CSC 173 Midterm Examination October 25th, 2005 Total: 75 points This is a closed book, closed notes exam. There are a total of 75 points. There is also an additional 5 points of extra credit (as part of your choice in question 5). In all cases, expl...
lecture-5.pdf
Path: Rochester >> CSC >> 458 Fall, 2009
Description: What is and Why Concurrency? What is a concurrent program? One with more than one active execution context (thread of control) Programming Models Standard models of parallelism shared memory (Pthreads) message passing (MPI) data parallel (Fort...
post-memory_model.pdf
Path: Rochester >> CSC >> 258 Fall, 2009
Description: Why Threads Cant be a library - and - The Java Memory Model CS 458 28 February 2007 Matt Post CSPicture@ University of Rochester 28 February 2006 Matt Post Big 2/458 @ University of Rochester 20 October 2007 Review Memory consistency models A...
BANSAL-FFX.pdf
Path: Rochester >> ECO >> 510 Fall, 2009
Description: An Exploration of the Forward Premium Puzzle in Currency Markets Ravi Bansal Duke University A standard empirical nding is that expected changes in exchange rates and interest rate differentials across countries are negatively related, implying that...
F97final.pdf
Path: Rochester >> ECO >> 108 Fall, 2009
Description: ECONOMICS 108 FINAL EXAM with ANSWERS Fall, 1997 There are 120 points on this exam, allocated as marked in parentheses after each problem. You have 3 hours. Print Name_ Sign Name__ ID Number_ Page 2_ Page 3_ Page 4_ Page 5_ Page 6_ Page 7...
m2f98ans.pdf
Path: Rochester >> ECO >> 108 Fall, 2009
Description: ECONOMICS 108, Fall, 1998 Second Midterm Professor Stockman There are 100 points on this exam. You have 55 minutes. Answers 1. Comment: \"A seller has an incentive to make its product fall apart quickly so that people have to replace it, raising d...
ssg.pdf
Path: Rochester >> ECO >> 108 Fall, 2009
Description: Is America Number 1? Study Guide Introduction Who\'s Number 1? What does it mean to be Number 1? You might say that your school football team is Number 1, meaning that the team usually beats other teams in its league. A Number 1 hit song sells the mo...
Home_2a.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 Homework 2 Suggested Answers PROBLEM 2.1 Consider three risky assets. The expected returns (in %) on those are: The variance-covariance matrix is For convenience you are also given the inverse variance-covariance matrix: a) Find the express...
Lect_9.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 9 The Arbitrage Pricing Theory, APT AIM OF LECTURE 9 Introduce the main arguments behind the APT. Learn how to form risk-free arbitrage portfolios. Understand how non-arbitrage arguments can pin down the structure of expected returns. Show ho...
Lect_10.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 10 Empirical Applications of the Capital Asset Pricing Model AIM OF LECTURE 10 Formulate different tests of the CAPM Become familiar with the literature testing the CAPM Form an understanding of the problems involved in empirical testing of t...
Lect_11.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 11 Empirical Applications of the Arbitrage Pricing Theory AIM OF LECTURE 11 Formulate different tests of the APT Become familiar with the literature testing the APT Form an understanding of the problems involved in empirical testing of the AP...
Lect_12.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 12 Introduction to Options AIM OF LECTURE 12 Learn the terminology of options Learn how to use payoff diagrams Introduce to Put-Call Parity 12.1 WHAT IS AN OPTION? An option is a contract (a financial contract) between two parties. We refer ...
Syllabus.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 FINANCIAL MARKETS: THEORIES & EVIDENCE Spring 2002 Time: Place: Instructor: Mondays and Wednesdays, 2pm to 3.15pm Dewey 2110D Thomas Renstrm Wallis Institute of Political Economy 110 Harkness Hall tel: 275-6834 e-mail: rnsm@troi.cc.rochester...
Lect_7.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 7 The Capital Asset Pricing Model AIM OF LECTURE 7 Derive the Capital Asset Pricing Model (CAPM) Become familiar with Fund-Separation results 7.1 OVERVIEW We will derive the (standard) Capital Asset Pricing Model (CAPM) through Mean-Variance...
Lect_8.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 8 Relaxing the assumptions: Zero-Beta CAPM, Taxation, and Borrowing-Lending constraints AIM OF LECTURE 8 Relax some of the assumptions underlying the Capital Asset Pricing Model (CAPM) 8.1 ZERO-BETA CAPM Why no risk-free asset? - inflation u...
Lect_6.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 6 The Investment Opportunity Set AIM OF LECTURE 6 Derive the Investment Opportunity Set Illustrate investment decisions of individuals 6.1 DERIVING THE INVESTMENT OPPORTUNITY SET In the last lecture we found the portfolios that give the mini...
Lect_13.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 13 Introduction to Continuous-Time Finance and Option Pricing AIM OF LECTURE 13 Become familiar with some continuous-time finance Learn how to form hedge portfolios with options Gain understanding of how a no-arbitrage argument underlies the ...
Rev_2.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Financial Markets: Theories and Evidence 1 ECO217 Revision Lecture 2 R2.1 RELAXING ASSUMPTIONS UNDERLYING THE CAPM No risk-free borrowing/lending The zero-beta CAPM [Black (1972)]. E[Rj] - E[Rz] = j.(E[Rm] - E[Rz]) where j COV(Rj,Rm)/VAR(Rm)...
Lect_3.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 3 Choice Under Uncertainty: Risk and Insurance Premia AIM OF LECTURE 3 Introduce the concept of risk aversion Provide measures of risk aversion Introducing the risk premium Solve for the Markowitz risk premium and analyze its components Intro...
Lect_2.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 2 Return, Risk and Choice Under Uncertainty AIM OF LECTURE 2 Calculate return and risk of portfolios using matrix algebra Revise portfolio formation in the context of short sales Revise expected utility and provide a numerical example Show th...
Samplee.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 Example of Final Examination for Financial Markets: Theories and Evidence INSTRUCTIONS Answer four questions of your choice (of the five below). Write your name clearly on all answer booklets. Pocket calculator is allowed. Formulae attached at...
Lect_5.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 5 The Efficient Frontier and MeanVariance Efficient Portfolios AIM OF LECTURE 5 Derive the Frontier Portfolios Analyse the properties of Frontier Portfolios 5.1 DERIVING THE FRONTIER PORTFOLIOS In the last lecture we formulated the optimisat...
Lect_4.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Lecture 4 Introduction to Mean-Variance Efficient Portfolios AIM OF LECTURE 4 Introduce the intuitive idea behind efficient portfolios Formulate the optimisation problem in order to find efficient portfolios Learn how to differentiate with respect to...
Sample3a.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 Answer to Sample Question 3 Question 3 (20%) Answer True or False (T or F ) T F (a) Two-fund separation is obtained only if there is a risk-free asset. (b) One-fund separation is obtained if returns are identically independently distributed. ...
Recit_1.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Recitation 1 Return, Risk and Choice Under Uncertainty INSTRUCTIONS Make an attempt to solve the problems below (you need not to hand in your answers). The TA will go through the problems in the recitation. You are expected to actively participate an...
Revision.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Financial Markets: Theories and Evidence 1 ECO217 Revision Lecture R.1 INTRODUCTION Course is about financial economics (pricing of risky assets) Three components: 1. 2. 3. Preferences (utility) Portfolios (the set of efficient portfolios is like a ...
Home_2.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Homework 2 Mean-Variance Efficient Portfolios INSTRUCTIONS 1. Solve the Problem 2.1 below (a-d), carefully showing your steps. 2. Hand in your work in the Class on Wednesday, March 20. 3. Your work will be marked and returned to you in respective rec...
Sampleq.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 Sample Questions for Financial Markets: Theories and Evidence AIM OF QUESTIONS These questions give you an idea of the level, structure and type of questions you may expect on the midterm examination (you will have to answer all three question...
Wrt_cred.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 Suggestion of Topics for Writing Credit INSTRUCTIONS Choose a topic and search for relevant literature. Produce an outline of your essay. Type your essay (5-10 pages), and hand it in on April 22. Prepare a presentation of your essay (10-15min)...
Home_1.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: Homework 1 Return, Risk and Choice Under Uncertainty INSTRUCTIONS 1. Solve the two problems below (taken from Lecture 2), carefully showing your steps. 2. Hand in your work in the Class on Wednesday, February 6. 3. Your work will be marked and return...
Sampleea.pdf
Path: Rochester >> ECO >> 217 Fall, 2009
Description: ECO217 Answer to Question 5A in Example of Final Examination Question 5 A Calculate the theoretical price of the a call option to buy one share. The current share price is 70, the exercise price is 65 and the annual risk free rate is 6%. The option m...
