Author Guidelines

Article/author guidelines can be found on the journal's own site here: http://www.actuarialsociety.org.za/News-and-Publications/Publications/South-African-Actuarial-Journal-671.aspx

Otherwise follow the guidlines below:

ABSTRACT
In this article instructions are given to authors of papers intended for publication in the South African Actuarial Journal (SAAJ) with regard to its editorial policy and the format required by SAAJ.

KEYWORDS
Publication; guidelines; South African Actuarial Journal

CONTACT DETAILS
Rob Rusconi, Tres Consulting, PO Box 951, Jukskeipark, 2153; Telephone: +27 11 4620149 (fixed) or +27 82 3345937 (mobile); E-mail: robr@tresconsulting.co.za

1. INTRODUCTION

1.1 In this article instructions are given to authors of papers intended for publication in the South African Actuarial Journal (SAAJ) with regard to its editorial policy and the format required by SAAJ. The latest version of this document is available on the SAAJ web page of the website of the Actuarial Society of South Africa (the Actuarial Society). ¹

1.2 The content of this article is arranged as follows. Section 2 explains the editorial policy of SAAJ. Authors are advised to read this before deciding whether to submit their articles to SAAJ. Section 3 gives details of the format required.

1.3 Authors requiring help with research methods are advised to refer to Guidelines on Research Methods, which is available on the web page of the Research Committee of the Actuarial Society. 2 It would be helpful to the Editor if authors complied with the standards of style applied by SAAJ as set out in Guidelines on Style for the South African Actuarial Journal, which is available on the SAAJ web page. This article itself complies with the format and style required.

2. EDITORIAL POLICY

2.1 SAAJ is published by the Actuarial Society. The Research Committee of the Actuarial Society is responsible to its Council for the publication of SAAJ. SAAJ is issued free to members of the Actuarial Society. It is also available for open access on the SAAJ web page of the website of the Actuarial Society. Its editorial policy is set and implemented by the Editor on the advice of an Editorial Advisory Panel, membership of which includes researchers of international standing in actuarial and related fields, as well as local actuaries who have experience in research. The President of the Actuarial Society is a member of the Editorial Advisory Panel ex officio. Other members of this Panel are appointed by the Editor on the advice of the Panel.

¹ www.actuarialsociety.org.za/Portals/2/Documents/InstructionsToAuthorsOfPapers-SAAJPublication- 20120215.pdf

² www.actuarialsociety.org.za/About/Committees/ResearchCommittee.aspx

SAAJ is published by the Actuarial Society, but it retains editorial independence. Nevertheless, sound working relationships are maintained between SAAJ and the Actuarial Society. The Editor is ex officio a member of the Research Committee of the Actuarial Society, to which he/she reports on matters relating to publication. Information regarding papers under consideration is reported confidentially. Rigorous editorial standards are maintained.

2.2 SAAJ publishes original, peer-reviewed research papers. The focus of SAAJ is on actuarial research—particularly, but not exclusively, on research of relevance to South Africa. The subject matter must, however, lie within the scope of actuarial work and be relevant and of interest to at least a minority of the profession in South Africa.

2.3 Papers must be original in order to be eligible for consideration. Useful introductions or reviews of subjects new to many actuaries, or useful descriptions of present practice or possible future practice will, however, be considered original if similar works are not available elsewhere. Such introductions, reviews or descriptions should be couched in language accessible to actuaries. The presentation of a paper at a conference or seminar, or the inclusion of a paper in the transactions of a conference, or the distribution of a paper by the authors to peers for the purposes of discussion prior to submission for publication, will not generally preclude it from being considered as an original paper. If, however, any part of a paper submitted for publication has been published or distributed in any manner, this should be drawn to the attention of the Editor. A paper submitted to SAAJ will be considered for possible publication only if the authors have certified that the paper in question is not under consideration by another journal, and will not be submitted to such a journal until and unless publication has been declined by SAAJ.

2.4 An empirical paper must contain, or permit reference to, sufficient detail of the methods and materials used in the study to permit replication in the hands of other researchers. No inconsistent data may be omitted and, except where clearly presented as being for illustrative purposes only, no fabricated data may be presented. The statistical treatment of data must be thorough and the conclusions reasonable. The existing relevant literature must be appropriately and fairly cited; in this respect, efforts should always be made to ensure that reference is made to the first report of a finding or conceptual insight rather than a later elaboration. Speculative deductions and postulations must be clearly specified and kept to a minimum.

2.5 Whilst priority is accorded from the date of acceptance of an article for publication, not from its date of receipt, both dates are always given in the published version.

2.6 The authors of a paper submitted for publication in SAAJ must declare in that paper any interest they may have in the outcome of the research and all funding sources must be acknowledged. If, in the opinion of the Editor, any such interest might have materially affected the outcome of the research in a manner that he or she considers inappropriate, or may appear to have so affected it, the Editor may decline publication of the paper. If the Editor becomes aware of such an undisclosed interest and the author refuses to disclose it, the Editor will decline publication of the paper.

2.7 Best practice is that studies addressing a particular question should not be broken up into a series of scattered short publications but preferably be presented once as a full record of the work and its results. In general, however, papers should be no longer than about 40 pages in SAAJ (about 16 000 words). Longer papers will be considered, but may have to be shortened, or to be restructured into two or more parts if they are otherwise suitable. Authors are advised to contact the Editor in advance if they suspect that their papers will exceed these limits.

2.8 It is the responsibility of the authors of a paper to clear copyright matters with the relevant parties. Immediately on the date of acceptance for publication by SAAJ, a paper becomessubject to the Actuarial Society’s Creative Commons attribution-only licence. In terms of this licence, the Actuarial Society, the author, or any third party is free to copy, distribute and transmit the paper, to adapt it and to make commercial use of it, provided the work is attributed to SAAJ. In that attribution the names of the author and SAAJ must be stated, together with the year of publication, the title of the paper and the volume of SAAJ in which it is to be published. If it has not yet been published, it must be shown as “forthcoming” in SAAJ. Further details of the Creative Commons licence may be found at the following web site address: www.creativecommons.org/licenses/by/3.0.

2.9 Papers must be submitted in English.

2.10 A paper may be submitted to the Editor by any author or group of authors, regardless of whether they are members of the Actuarial Society.

2.11 On receipt of a paper the Editor decides whether it should be considered for possible publication. If so, the Editor circulates to the Editorial Advisory Panel a copy of the title and abstract of the paper, calling for nominations of scrutineers. If not, the Editor declines publication. For the purposes of consideration for possible publication the paper is scrutinised by at least two scrutineers nominated by the Editorial Advisory Panel or selected by the Editor, and appointed by the Editor. Any person so nominated or selected should, to the knowledge of the Editor and the Panel member making the nomination, be a scholar, or a practitioner of the actuarial profession or another learned profession, who has not previously co-published extensively with the authors, who is free of known bias in relation to the subject matter, the authors and their employers and clients, and who can cover, from a position of expertise or experience, the topics dealt with in the paper. If a paper requires expertise or experience in more than one topic, a Panel member may nominate a scrutineer with expertise or experience in one such topic, provided that the nomination must be qualified accordingly and that the Editor must ensure that other scrutineers are appointed with expertise or experience in the other such topics. If a scrutineer so appointed has, or may reasonably be considered to have, any interest in the outcome of the research, or if he/she does not satisfy the requirements of this paragraph, he/she is required either to decline appointment or to inform the Editor, who will reconsider the appointment in the light of this information.

2.12 The names of the scrutineers are not disclosed to the authors unless a scrutineer wishes to be known. Scrutineers will not enter into any correspondence or discussion with the authors or with other scrutineers with regard to the paper except via the Editor. Authors are advised not to nominate scrutineers. Further details of the scrutiny process are set out in Guidelines to Scrutineers of a Paper Submitted for Publication in the South African Actuarial Journal, which is available on the SAAJ web page.

2.13 If a paper authored or co-authored by the Editor or by an author with whom the Editor is associated by extensive co-publication, employment or common clients is to be submitted for publication in SAAJ, the Editor requests the Editorial Advisory Panel to nominate an acting editor, who will act as the editor for the purpose of consideration of that paper. If a paper authored or co-authored by a member of the Panel or by an author with whom such a member is associated by extensive co-publication, employment or common clients is submitted for publication in SAAJ, that member will not participate in the process of consideration of that paper.

2.14 Authors should not rely on the scrutineers to check the accuracy of any numerical calculations whose results are shown in the paper. Notwithstanding any alterations made on the recommendations of scrutineers or the Editor, authors remain solely responsible for the accuracy of the contents of their papers. Any error or falsification detected after the publication of a paper must be retracted in a subsequent issue of SAAJ.

2.15 The Editor assesses the reports of the scrutineers to decide whether, individually and collectively, they constitute the basis for the publication of the article in question. On the advice of the scrutineers, the Editor decides whether:
(a) the paper is suitable for publication, subject to certain amendments required or suggested by the scrutineers (in which case the scrutineers so recommending are requested to indicate the amendments required with reference to the paragraph numbers concerned; on receipt of an amended paper the Editor satisfies himself/herself that the amendments have been adequately dealt with, for which purpose he/she may refer particular amendments to those scrutineers for further advice);
(b) the paper needs major revision before it can be reconsidered (in which case, with reference to the paragraph numbers concerned, the scrutineers so recommending are requested to outline the revision required and, at their discretion, any minor amendments required; on receipt of a revised paper it is referred to those scrutineers for reconsideration);
(c) the paper is unsuitable for SAAJ, and should be submitted to an alternative publication (in which case the scrutineer so recommending is requested to suggest an alternative publication or type of publication); or
(d) publication of the paper should be declined.

Acceptance without amendment is unlikely. The decision of the Editor is final.

2.16 Authors should be advised within three months of receipt of a paper by the Editor whether it is accepted for publication and if so what amendments or revision are required. However, it is sometimes difficult to find appropriate scrutineers and, although, once appointed, scrutineers are asked to make their comments within about six weeks, or at most two months, it is not always possible for them to do so. When an amended or revised paper is received, the Editor reviews the amendments made in response to the reports of those scrutineers who advised that the paper be amended. When a revised version of the paper is received by the Editor, copies of this are sent to those scrutineers who advised that the paper should be substantially revised and the process is repeated.

2.17 When the authors receive a report from the scrutineers, they should ensure that each of the scrutineer’s comments is attended to. If it is not obvious how the authors have amended the paper in response to a particular comment made by a scrutineer, they should explain their amendments in a covering note to the Editor with reference both to the paragraph numbers involved and (where applicable) to the scrutineer’s comment number. The authors may defend the paper against comments made by a scrutineer. However, it is recommended that any such defence should generally be made in the paper rather than in a note to the Editor. This may entail explaining both the criticism and the defence in the paper. Such defences will generally enhance the quality of the paper. It should be borne in mind that the scrutineers have been chosen as being particularly well versed in the subject of the paper. If they appear to have misunderstood something in the paper, it is quite likely that other readers will do so too. Care should therefore be taken to avoid such misunderstandings by suitable amendment of the paper. Authors should not make substantial or extensive amendments to the paper other than those required by the scrutineers or the Editor. If any criticism is unclear, the authors should query it with the Editor before resubmitting the paper.

2.18 After acceptance the authors see one set of proofs for checking, subject to a deadline indicated by the Editor. This will generally be about one week. Authors who are not members of the Actuarial Society will each receive one copy of the issue in which their article is published.

2.19 Authors of papers read at conventions or sessional meetings of the Actuarial Society are invited to submit them (after amendment, if necessary to conform to these instructions and to the Guidelines on Style) for publication in SAAJ. A reader who wishes to respond to such a paper by means of a letter to the Editor may inform the Editor accordingly. Once a paper has been accepted for publication in SAAJ, it is placed on SAAJ’s web page. Unless specifically requested by the respondents, responses to a paper will not themselves be refereed, but the authors of the original paper will have the right to reply in the same issue. In order to avoid delays in the publication of SAAJ, it may be necessary to apply time constraints to readers’ responses and authors’ rights to reply. Publication of responses to papers will in any case be at the discretion of the Editor on the advice of the Editorial Advisory Panel.

2.20 Besides peer-reviewed papers, SAAJ may include:
 an editorial;
 letters to the Editor;
 abstracts of recent research reports, dissertations and theses for postgraduate degrees at South African universities;
 abstracts of monographs and other publications on South African matters of actuarial interest;
 abstracts of articles of interest to actuaries in other South African journals;
 abstracts of articles of interest to South African actuaries in other journals and publications;
 book reviews;
 corrigenda; and
 a cumulative index to articles in SAAJ.

Readers are welcome to submit letters to the Editor or suggestions of books for review or items for inclusion in the abstracts.

2.21 Copies of forthcoming issues may be ordered from the Actuarial Society at PO Box 4464, Cape Town 8000. Limited numbers of copies will also be retained by the Actuarial Society for subsequent sale. All correspondence with regard to editorial policy and the publication of papers should be addressed to the Editor, Rob Rusconi, whose contact details appear at the start of this article.

3. FORMAT
3.1 GENERAL REMARKS
3.1.1 This section gives guidance on format to authors, or to their assistants, who are preparing papers for publication in SAAJ. If they are followed, then the typesetting process will be made much easier. This document is laid out in the format that should be followed. Authors are not required to conform in detail to this format, except as described in this section; the format is applied during the typesetting process. In particular, conventions with regard to line spaces and indenting, and variations between bold and plain typefaces, need not be adhered to by authors. Authors are requested to avoid the use of unnecessary formatting information. In particular, indenting should be applied by means of tabs, not multiple spaces.
3.1.2 It is assumed that authors will prepare text using a word-processing system on a computer, and are able to submit the text by e-mail. The typesetter can read files prepared by several different word-processing systems, so it is convenient if the text is laid out in the required format. If the authors wish to prepare text on a system that may not be readable by the Editor, this should be discussed with the Editor before submission.
3.1.3 Text is prepared on A4 paper, as this is, although it is altered during typesetting to fit on the A5 size of SAAJ. This document uses the Times New Roman font with a normal font size of 12 point, with single line spacing; authors are requested to follow this format.
3.1.4 Arrangements can be made for the editing of the format of a paper by the typesetter to conform to the requirements of SAAJ before it is submitted. The contact details of the typesetter may be obtained from the Editor. It should be noted, however, that any such arrangements are to be made between the authors and the typesetter, that the costs will be borne by the authors and that the Actuarial Society will not be involved.

3.2 LAYOUT
3.2.1 INTRODUCTION
Each paper begins with:
– title;
– authors’ names;
– abstract;
– keywords;
– contact details; and
– quotation (if desired).

These are laid out as shown at the top of this document, and as described below. The Editor will insert the dates of submission and acceptance for publication.

3.2.2 TITLE OF THE PAPER
The title should not include abbreviations.

3.2.3 AUTHORS’ NAMES
3.2.3.1 The sequence is: “By”; initials; surname. No qualifications are given. There are no spaces or points between initials.
3.2.3.2 If applicable, the names of the principal authors should be shown first; the names of other authors should be shown only if they have made substantial intellectual or conceptual contributions to the research or to the drafting or revision of the intellectual content of the paper, and if they share responsibility for the paper. Contributions by other persons should be acknowledged as such (cf. ¶3.5 below).

3.2.4 ABSTRACT
3.2.4.1 Every paper requires an abstract; this should be a summary of the paper, of not more than 250 words, preferably in one paragraph. The abstract is often copied by other journals that print lists of papers with abstracts, so it should be written in a way that will encourage others to read the paper. It is preceded by the heading ‘ABSTRACT’.
3.2.4.2 Only paragraphs after the first are indented.

3.2.5 KEYWORDS
3.2.5.1 Next come keywords; these are also used by other journals and in citation indices, so words should be chosen that give an indication of what is of interest in the paper. Not more than about five keywords should normally be chosen. The list of keywords is preceded by the heading ‘KEYWORDS’.
3.2.5.2 The keywords themselves are not indented. Only the first has an initial capital letter. They are separated by semi-colons (;) followed by one space. There is no point (.) after the last keyword.

3.2.6 CONTACT DETAILS
The author’s contact details (or if there are more than one, a contact author with whom the Editor and readers may correspond) are shown under the heading ‘CONTACT DETAILS’. The details to be given are: title, given name, surname, postal address (including country if not in South Africa), all separated by commas, followed by ‘; Tel:’, the telephone number preceded by the country code if not in South Africa and the area code; ‘; Fax:’, the fax number similarly stated; ‘E-mail:’ and the e-mail address.

3.2.7 QUOTATION Some authors like to include a quotation at the start of their papers. This is acceptable. If it is done, the quotation comes next, followed by the name of the author and possibly the source of the quotation. There is no point (.) at the end.

3.3 THE MAIN TEXT
3.3.1 The main text of the paper comes next. This is always separated into sections, numbered 1, 2 etc. Each section may be subdivided into subsections, numbered 1.1, 1.2 etc., and these in turn may be subdivided into sub-subsections, numbered 1.1.1, 1.1.2 etc. Finally there are paragraphs, which may be numbered 1.1, 1.2 etc. or 1.1.1, 1.1.2 etc., or 1.1.1.1, 1.1.1.2 etc., depending on the level of subsections needed. Never go beyond four levels of numbering.
3.3.2 Authors may choose to number paragraphs manually, but they are welcome also to use their word-processor’s automatic numbering feature, taking care to follow the formatting specification set out in the paragraphs following this one.
3.3.3 A section is headed with the section number (e.g. 1), followed by a point (.), then a tab space, then the title of the section, in full capitals bold-face.
3.3.4 A subsection starts with its number (e.g. 1.1), followed by a tab space (no point), then the subsection title, in full capitals regular weight.
3.3.5 A sub-subsection starts with its number (e.g. 1.1.1), followed by a tab space (no point), then the sub-subsection title in small capitals regular weight, with full capitals for the initial letters of important words.
3.3.6 A paragraph under a major section heading starts with its number (e.g. 1.1), followed by one tab space (no point), then the text. A paragraph in a subsection or subsubsection is indented one tab space, then starts with its number (e.g. 1.2.3 or 1.2.3.4 as appropriate), followed by one tab space (no point), then the text. Avoid using a new paragraph for each sentence, except where appropriate.
3.3.7 If a section (including a subsection or a sub-subsection) contains only one paragraph, that paragraph does not need a number.
3.3.8 There is often a need to refer to another section or paragraph in the text. In this case refer to sections, subsections and sub-subsections all as, e.g. ‘section 1’, ‘section 2.1’ or ‘sections 3.2.1 and 3.2.2’; refer to paragraphs as, e.g. ¶4.1, ¶¶4.2.1 to 4.2.4 etc. Put ‘paragraph’ if you do not have the symbol ¶ available.
3.3.9 There should be only one space between sentences. Similarly, (except for emdashes ‘—’ and ellipses ‘…’) there should be a space—and only one space—after a punctuation mark and none before it. However, do not put any spaces after the last point in a paragraph. 3.3.10 Except for the referencing of unpublished sources (see ¶3.6.3), footnotes should not be used. Endnotes should not be used.
3.3.11 Formulae should generally be placed centrally. Exceptions may occur where a series of equations would be better aligned at the ‘=’ sign, especially where the left-hand side of the equations is omitted after the first, or where a list of definitions is given in the form of equations after a centrally placed equation. Formulae referred to in the text should be numbered in brackets in simple sequence from the start to the end of the paper using a right-justified tab to the right of the formulae. Formulae not referred to in the text should not be numbered. An equation editor should be used for formulae.
3.3.12 Italics should be avoided except to denote foreign words normally italicised in English usage or titles of publications. In particular, the use of italics or bold typeface for emphasis should be avoided.

3.4 POINTS
3.4.1 Sometimes items are best set out as a series of points. These can be introduced with a series of dashes (–) or where it is necessary to refer to them individually (as in ¶¶2.15 and 3.6.2), with (1), (2) etc. or (a), (b) etc. (but not with ‘bullets’ (•)). In either case they are fully justified to the left margin, i.e. not indented, unless a series of sub-points come within a point, in which case they should be indented one tab space. The text within a point should be indented and aligned. See ¶2.20 for an example.
3.4.2 A list of points should be introduced with a colon (:) at the end of the preceding line. If the text in each point is short, each should start with a small letter, and should end with a semi-colon (;), except the second-last, which should end with ‘; and’, and the last, which should end with a point (.). Each point should be grammatically consistent with the introduction, and grammatically parallel to the other points, so that the whole reads as one sentence. Points should not commence with capital letters unless they are sentences. If the text in each point is longer, amounting to one or more sentences, then each should start with a capital and end with a point, and should be structured as a complete sentence. Point format should not be used for a series of substantial paragraphs. A blank line should be inserted at the end of a list of points.

3.5 ACKNOWLEDGEMENTS
3.5.1 Some authors like to include acknowledgements to those who have helped them in the development of their paper. This should appear at the end of the main text, under the heading ‘ACKNOWLEDGEMENTS’. They should not be too fulsome.
3.5.2 As indicated in ¶2.6, all funding sources must be acknowledged.

3.6 REFERENCES
3.6.1 All work done by other authors and used or referred to in the text should be attributed to the authors and every statement obtained from another source should be properly cited. No statement that has already been made by another author should be made in the text without citing that author. The purposes of these requirements are to avoid plagiarism, to establish the authority of the statements made and to avoid covering ground that has already been covered in the literature. (Every statement made must have some source of authority, which should be clear to the reader. The authority may be from observed data, from logical argument or inference, or from the literature.) An exception to these requirements is knowledge that is fundamental to actuarial science or to secondary education. If an entire section is based on a particular author’s work, this should be made clear. Where it is appropriate to quote the actual words of another author, they should be shown accordingly (see ¶4.2 of Guidelines on Style). Extensive quotations may require permission from the authors concerned; this is the responsibility of the authors of the paper.
3.6.2 A document, whether printed or retrieved from the internet, should be treated as published literature if and only if:
(a) it has been published (i.e. been made available to the public, either at reasonable cost or free of charge) by a stated publisher other than the author; and
(b) the year in which it was published is stated by the publisher; and
(c) (i) it is published in a peer-reviewed journal; or
(ii) it is published by an authoritative publishing house; or
(iii) there is other reason to consider it authoritative.

3.6.3 For the avoidance of doubt and in line with the principles set out herein, Wikipedia and similar sources should never be cited as a source of information. Wikipedia is authored by anonymous contributors who endeavour to provide original references. Authors should always follow the primary sources that Wikipedia contributors provide.
3.6.4 A document that has been accepted for publication as published literature may be cited as forthcoming. An article submitted for publication but not yet accepted should be treated as unpublished. A document should be listed in the references if and only if it is cited as a source of authority. The listing of references is explained in ¶¶3.6.7–14 below. If a document is cited merely as a source of information, particularly in the case of a document that is not published in a peer-reviewed journal or by an authoritative publishing house, it should be cited merely by means of a footnote; it is recommended that the word-processor’s footnote system be used for this purpose. Where possible, published works should be cited in preference to unpublished works. A statute should be referred to as, e.g., ‘the Pension Funds Act, 1956, 3 with a footnote giving its full name and number and except in the case of the Parliament of South Africa, the legislature.
3.6.5 In the text, except where the citation itself is in parentheses, it should be in the form: ‘Smith (1995)’ or, if it is an unpublished document, ‘Brown (unpublished)’ or, if it has been accepted for publication: ‘Malan (forthcoming)’. If there are two authors, their names are joined by ‘&’, not ‘and’, e.g. ‘Van der Merwe & Nkomo (1996)’. If there are three authors put e.g. ‘Smith, Van der Merwe & Nkomo (1997)’. If there are more than three, put e.g. ‘Smith et al. (1998)’. If you refer to more than one paper by one author with the same year, or unpublished or forthcoming, put e.g. ‘Smith (1998a)’, ‘Smith (1998b)’ etc. or ‘Brown (unpublished a)’, Brown ‘(unpublished b)’ etc.. If the citation itself is in parentheses, put e.g. ‘(Smith, 1995)’. Such a citation may be used after a sentence or at the end of a paragraph that is attributed to the author concerned. If you need to refer frequently to the same work and the citation is lengthy, put e.g. ‘Smith, Van der Merwe & Nkomo (1997) (referred to as SVN)’, and refer to it thereafter as ‘SVN’. Instead of referring in full to the work last cited, ‘ibid.’ may be inserted in brackets. Instead of inserting the date in a work already cited, ‘op. cit.’ may be substituted, e.g. ‘Smith (op. cit.)’ or ‘(Smith, op. cit.)’ provided there is no possibility of confusion. Citations in the text are taken as referring to the authors, and not to the work, unless the context implies otherwise. 3 Act no. 24 of 1956 as amended, Republic of South Africa Personal pronouns referring to them should be chosen accordingly, and where there is more than one author, verbs following them should be in the plural. Citations such as ‘Smith’s (1995) model’ may also be used.
3.6.6 In the citation, reference may be made to the pages referred to by showing them after a colon within the bracket, e.g. Smith (1998: 276–78) or (Smith, 1998: 276–78).
3.6.7 All citations to the literature in the text of the paper (including appendices) should be listed as references at the end of the main text, but before appendices. This list is not intended to be a bibliography of sources that the authors have consulted, nor of sources that the reader might wish to consult; works not cited in the text should not be listed here. References are placed in alphabetical order by author, and by date order for any one author.
3.6.8 The list of references should be headed ‘REFERENCES’.
3.6.9 Each reference should be given in the required format. Each begins with the surname of the author, followed by a comma and a space, followed by initials (not full forenames), with no spaces or points between. If there is more than one name, each is put in the same form, with ‘&’ between the last two, and a comma between earlier names. Put all the names, however many there are. Some publications do not have an obvious author; in that case the name of a committee or government department will do, but it should be the same name as used in the main text.
3.6.10 After the authors’ names comes the year of publication, in full, in parentheses, e.g. ‘(1999)’ or, in the case of a forthcoming or unpublished article, ‘(forthcoming)’ or ‘(unpublished)’. The year etc. may be followed by a small letter to distinguish works by the same author in the same year, e.g. ‘(1998a)’, ‘(1998b)’, ‘(unpublished a)’ etc. The closing parenthesis is followed by a point and a space.
3.6.11 Next comes the title of the work referred to. The style of this depends on what it is. If it is a book, the title is put in italics, with initial capitals for important words, and this is followed by a point and one space. After this come the name of the publisher, a comma, and the town of publication, in roman type.
3.6.12 If the work referred to is an article in a journal, the title of the article is in roman type, with initial capitals only for the first word, followed by a point and one space. Then comes the name of the journal in which it appears, spelled out in full, in italics and in title case. The name of the journal is followed by a space, then the volume number in bold-face and the issue number in regular weight in brackets, then another comma and a space, then the first page number. Include the last page number too (excluding leading digits corresponding to those of the first), with an en-dash (–) separating the numbers.
3.6.13 For an unpublished work, the title is printed in regular (not italic) font. For a thesis or dissertation, the details should be shown as e.g. ‘Unpublished PhD thesis, Department of Actuarial Science, University of ….’ If a year of authorship is indicated, this should be shown. As indicated above, an article submitted for publication but not yet accepted should be treated as unpublished; the name of the journal should not be stated.
3.6.14 For a document retrieved from the internet, the format is the same as for a printed document, except that the name of the publisher should be followed by the uniform resource locator (URL) of the document in the conventional format. If, however, that URL is unnecessary or excessively long, the URL of the web page from which the document may be accessed is sufficient. This is followed by the date on which the information was retrieved, e.g. ‘retrieved 6 June 2008)’. If the document is also published in hard copy it is not necessary to show these details. If to your knowledge the document has been assigned a digital object identifier (DOI), this should be shown.
3.6.15 There are other sorts of references, and they need to be laid out appropriately. The purpose is to allow the interested reader to find the piece referred to, for example in a library or on the internet, so enough information must be given for it to be found, but not more. For illustrative purposes, references of various types are set out in Appendix A in the format required.

3.7 APPENDICES Often it is appropriate to put supplementary material, series of tables or figures, data sources, formulae, proofs etc. into separate appendices. These follow the main text (after acknowledgements and the list of references). If there is only one appendix it is labelled Appendix”. If there are more than one they are labelled Appendix A, Appendix B etc. An appendix is treated like a section of the paper; subsections and paragraphs are labelled, e.g. A.1, A.2., A.3 etc. Each appendix starts on a new page, and is headed, e.g. APPENDIX A, in capitals, with the title of the appendix in a similar style below it.

3.8 TABLES
3.8.1 Many papers include tables of various kinds. Tables should be numbered and headed, so that they can be moved to an appropriate place in the text. They should normally be placed just after the paragraph in which they are first referred to, or at the next convenient point thereafter, which may be at the top of the next page. A long series of tables is best put into an appendix.
3.8.2 Tables are numbered as Table 1, Table 2 etc, consecutively throughout the text. A large table that has to be split into several sections might be numbered e.g. Table 3a, Table 3b etc. Tables in appendices should be labelled, e.g. Table C.1 etc..
3.8.3 Tables should generally be introduced, explained and discussed in the text. Tables are referred to in the text as, e.g. Table 1 or Tables 1 to 6, with an initial capital T.
3.8.4 A numbered table has a heading, placed above it, that starts e.g. Table 3, followed by a point, then one space, then the title of the table, with an initial capital, but no other capitals, and with no final point. It is better if a table can fit vertically on the page (in portrait style), but a wide table may need to go sideways (in landscape style). The authors should enter each table in the document where it is to appear. The word-processor’s tabling system may be used for this purpose. It is not necessary for the author to be concerned about the layout of tables; this will be seen to during the typesetting process. It must be clear, however, which headings and sub-headings relate to each column or row of entries.
3.8.5 Text in the table should be in 12-point roman type.

3.9 FIGURES
3.9.1 Many papers require figures, diagrams or graphs. These are usually referred to as ‘figures’. Similar comments apply to figures as to tables. Often figures are graphs, drawn by means of a graph plotting system. No colour should be used. The author should paste figures electronically into the document. If the figure is prepared electronically by means of a standard system, it should be included in a separate file. (A spreadsheet file may be used for this purpose. Unnecessary data should not be included.)
3.9.2 A figure should be designed to give the maximum of information with the minimum of print. A good book on how to design graphs is Tufte (1983). In general, use only left-hand and bottom axes; make sure that both are appropriately labelled. Use a vertical logarithmic scale when it is more suitable to do so (e.g. for mortality rates or share prices).
3.9.3 Figures should be numbered as Figure 1, Figure 2 etc. consecutively throughout the text. Figures in appendices should be labelled, e.g. Figure D.1, Figure D.2 etc. within Appendix D.
3.9.4 Figures should generally be introduced, explained and discussed in the text. Figures are referred to in the text as, e.g. Figure 4, or Figures 1 to 6, with initial capital F.
3.9.5 A numbered figure has a caption, placed below it, in roman type, starting e.g. Figure 4, followed by a point, then one space, then the title of the figure, with an initial capital, but no other capitals, and with no final point. For ease of identification, the figure number and caption should be shown with the figure on the worksheet of the separate electronic file used for it (cf. ¶3.9.1).
3.9.6 Text in the figure should be in 12-point roman type.

3.10 RUNNING HEADINGS, FOOTERS AND PAGE NUMBERS Running headings and footers should not be used. The pages should be numbered in the top right-hand corner using the author’s word-processing system.

3.11 FORM AND DATE OF SUBMISSION The paper should be submitted to the Editor in electronic form either by e-mail or by post. Subject to alternative arrangements with the Editor, a paper for inclusion in a particular year’s volume should reach the author by 31 December the previous year. Papers will, however, be received for possible publication at any time of the year. A paper accepted for publication will be published in the first issue after its acceptance. Publication of an issue will not be delayed for the purpose of including a paper received after 31 December the previous year.

ACKNOWLEDGEMENTS The author acknowledges the help of Mrs AD Hart, on whose work section 3 is based. The contribution of Professor AD Wilkie is also acknowledged. Ms J Friedlander’s continuing contribution to the formatting and typesetting standards of SAAJ is appreciated.

REFERENCES Tufte, ER (1983). The Visual Display of Quantitative Information. Graphics Press, Cheshire, Conn.

APPENDIX A

HYPOTHETICAL REFERENCES FOR ILLUSTRATIVE PURPOSES

A set of hypothetical references is set out below for illustrative purposes. This includes: – unpublished material: Brown (unpublished a), an article on the Internet, Brown (unpublished b), a paper distributed at a convention of the Actuarial Society, and Brown (unpublished c), a document distributed by a firm of consultants;
– a paper accepted for publication in a journal: Malan (forthcoming);
– journal articles: Smith (1995a, 1995b) and Smith et al. (1997); – a dissertation: Smith (unpublished);
– a chapter (Smith et al., 1998) in an edited book (Davidson & Henderson, 1998); and
– a book: Van der Merwe & Nkomo (1996).

REFERENCES
Brown, JD (unpublished a). Modelling the actuary. Actuarial Society of South Africa, http://www.cgi.org.za/downloads/brown.pdf, 3/6/2008
Brown, JD (unpublished b). The role of actuaries in healthcare. Convention, Actuarial Society of South Africa, 2000
Brown, JD (unpublished c). The WW survey of South African insurers, 2000. Unpublished document, White & Wong Financial Consultants Ltd, Johannesburg
Davidson, AL & Henderson, RB (eds.) (1998). Social Security in Africa, 2nd edition, revised by EF Molefe. Baobab Press, Pretoria
Malan, FK (forthcoming). AIDS mortality in South Africa. Forthcoming in SAAJ
Smith, JC (1995a). The use of models in actuarial science. European Actuarial Journal 12(1), 113–38 dx.doi.org/10.5223/eaj.v12i1.3
Smith, JC (1995b). On the analysis of surplus of defined-benefit pension funds. Transactions of the Actuarial Society of South Africa 5, 58–73
Smith, JC (unpublished). Benefit targeting in defined-contribution retirement funds. Unpublished MSc dissertation, School of Statistics and Actuarial Science, University of the Witwatersrand, Johannesburg, 1997
Smith, JC, Jackson, RN, Anderson, VA & Williamson, DK (1998). A history of social security in South Africa. In Davidson & Henderson (1998: 135–79)
Smith, JC, Van der Merwe, WL & Nkomo, PJ (1997). The role of actuaries in the creation of financial security. South African Actuarial Journal 4, 143–82 dx.doi.org/10.5223/saaj.v4.2
Van der Merwe, WL & Nkomo, PJ (1996). The Population of South Africa. Jackal Press, Johannesburg