Chapter 3. Systems of Linear Equations: Geometry

Transcription

1 Chapter 3 Systems of Linear Equations: Geometry

2 Motiation We ant to think about the algebra in linear algebra (systems of equations and their solution sets) in terms of geometry (points, lines, planes, etc). x 3y = 3 2x + y = 8 This ill gie us better insight into the properties of systems of equations and their solution sets. Remember: I expect you to be able to dra pictures!

3 Section 3.1 Vectors

4 Points and Vectors We hae been draing elements of R n as points in the line, plane, space, etc. We can also dra them as arros. Definition A point is an element of R n, dran as a point (a dot). the point (1, 3) A ector is an element of R n, dran as an arro. When e think of an element of R n as a ector, e ll usually rite it ectically, like a matrix ith one column: ( ) 1 =. 3 [interactie] the ector ( 1) 3 The difference is purely psychological: points and ectors are just lists of numbers.

5 Points and Vectors So hy make the distinction? A ector need not start at the origin: it can be located anyhere! In other ords, an arro is determined by its length and its direction, not by its location. These arros all represent the ector ( 1 2). Hoeer, unless otherise specified, e ll assume a ector starts at the origin.

6 Vector Algebra Definition We can add to ectors together: a x a + x b + y = b + y. c z c + z We can multiply, or scale, a ector by a real number c: x c x c y = c y. z c z We call c a scalar to distinguish it from a ector. If is a ector and c is a scalar, c is called a scalar multiple of. (And likeise for ectors of length n.) For instance, = 7 and 2 2 =

7 Vector Addition and Subtraction: Geometry 5 = = The parallelogram la for ector addition Geometrically, the sum of to ectors, is obtained as follos: place the tail of at the head of. Then + is the ector hose tail is the tail of and hose head is the head of. Doing this both ays creates a parallelogram. For example, ( 1 3) + ( 4 2) = ( 5 5). 5 = = Why? The idth of + is the sum of the idths, and likeise ith the heights. [interactie] Vector subtraction Geometrically, the difference of to ectors, is obtained as follos: place the tail of and at the same point. Then is the ector from the head of to the head of. For example, ( 1 4) ( 4 2) = ( 3 2 ). Why? If you add to, you get. [interactie] This orks in higher dimensions too!

8 Scalar Multiplication: Geometry Scalar multiples of a ector These hae the same direction but a different length. Some multiples of ( ) 1 = 2 ( ) 2 2 = 4 1 ( ) 1 2 = 2 1 ( ) 0 0 = 0 All multiples of. [interactie] So the scalar multiples of form a line.

9 Linear Combinations We can add and scalar multiply in the same equation: = c c c p p here c 1, c 2,..., c p are scalars, 1, 2,..., p are ectors in R n, and is a ector in R n. Definition We call a linear combination of the ectors 1, 2,..., p. The scalars c 1, c 2,..., c p are called the eights or coefficients. Example ( ( 1 1 Let = and =. 2) 0) What are some linear combinations of and? [interactie: 2 ectors] [interactie: 3 ectors]

10 Poll Poll Is there any ector in R 2 combination of and? that is not a linear No: in fact, eery ector in R 2 is a combination of and. (The purple lines are to help measure ho much of and you need to get to a gien point.)

11 More Examples ( 2 What are some linear combinations of =? 1) What are all linear combinations of? All ectors c for c a real number. I.e., all scalar multiples of. These form a line. Question What are all linear combinations of ( ( ) 2 1 = and =? 2) 1 Anser: The line hich contains both ectors. What s different about this example and the one on the poll? [interactie]

12 Section 3.2 Vector Equations and Spans

13 Systems of Linear Equations Question Is 16 a linear combination of 2 and 2? This means: can e sole the equation x 2 + y 2 = here x and y are the unknons (the coefficients)? Rerite: x y 8 x y 8 2x + 2y = 16 or 2x 2y = 16. 6x y 3 6x y 3 This is just a system of linear equations: x y = 8 2x 2y = 16 6x y = 3.

14 Systems of Linear Equations Continued Conclusion: x y = 8 2x 2y = 16 6x y = 3 matrix form ro reduce solution = [interactie] (this is the picture of a consistent linear system) x = 1 y = 9 What is the relationship beteen the original ectors and the matrix form of the linear equation? They hae the same columns! Shortcut: You can make the augmented matrix ithout riting don the system of linear equations first.

15 Vector Equations and Linear Equations Summary The ector equation x x x p p = b, here 1, 2,..., p, b are ectors in R n and x 1, x 2,..., x p are scalars, has the same solution set as the linear system ith augmented matrix 1 2 p b, here the i s and b are the columns of the matrix. So e no hae (at least) to equialent ays of thinking about linear systems of equations: 1. Augmented matrices. 2. Linear combinations of ectors (ector equations). The last one is more geometric in nature.

16 Span It is important to kno hat are all linear combinations of a set of ectors 1, 2,..., p in R n : it s exactly the collection of all b in R n such that the ector equation (in the unknons x 1, x 2,..., x p) has a solution (i.e., is consistent). x x x p p = b the set of such that Definition Let 1, 2,..., p be ectors in R n. The span of 1, 2,..., p is the collection of all linear combinations of 1, 2,..., p, and is denoted Span{ 1, 2,..., p}. In symbols: Span{ 1, 2,..., p} = { x x x p p x 1, x 2,..., x p in R }. Synonyms: Span{ 1, 2,..., p} is the subset spanned by or generated by 1, 2,..., p. This is the first of seeral definitions in this class that you simply must learn. I ill gie you other ays to think about Span, and ays to dra pictures, but this is the definition. Haing a ague idea hat Span means ill not help you sole any exam problems!

17 Span Continued No e hae seeral equialent ays of making the same statement: 1. A ector b is in the span of 1, 2,..., p. 2. The ector equation has a solution. x x x p p = b 3. The linear system ith augmented matrix 1 2 p b is consistent. [interactie example] (this is the picture of an inconsistent linear system) Note: equialent means that, for any gien list of ectors 1, 2,..., p, b, either all three statements are true, or all three statements are false.

18 Pictures of Span Draing a picture of Span{ 1, 2,..., p} is the same as draing a picture of all linear combinations of 1, 2,..., p. Span{} Span{, } Span{, } [interactie: span of to ectors in R 2 ]

19 Pictures of Span In R 3 Span{} Span{, } Span{u,, } Span{u,, } u u [interactie: span of to ectors in R 3 ] [interactie: span of three ectors in R 3 ]

20 Obseration about spans 0 Ho many ectors are in Span 0? 0 A. Zero B. One C. Infinity In general, it appears that Span{ 1, 2,..., p} is the smallest linear space (line, plane, etc.) containing the origin and all of the ectors 1, 2,..., p. We ill make this precise later.

21 Summary The hole lecture as about draing pictures of systems of linear equations. Points and ectors are to ays of draing elements of R n. Vectors are dran as arros. Vector addition, subtraction, and scalar multiplication hae geometric interpretations. A linear combination is a sum of scalar multiples of ectors. This is also a geometric construction, hich leads to lots of pretty pictures. The span of a set of ectors is the set of all linear combinations of those ectors. It is also fun to dra. A system of linear equations is equialent to a ector equation, here the unknons are the coefficients of a linear combination.